Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

Soft Matter
View Article Online
PAPER View Journal | View Issue

Poly(vinyl alcohol)-induced thixotropy of an

L-carnosine-based cytocompatible, tripeptidic
Cite this: Soft Matter, 2019,
15, 433 hydrogel†
a
Rita Das Mahapatra, Joykrishna Dey *a and Richard G. Weiss b

The generally poor mechanical stability of hydrogels limits their use as functional materials for many
biomedical applications. In this work, a poly(vinyl alcohol) (PVA) embedded hybrid hydrogel of a b-amino
acid-containing Fmoc-protected tripeptide was produced at physiological pH (7.4) and room tempe-
rature. The hydrogel system was characterized by a number of techniques, including UV-vis, fluorescence,
circular dichroism, FT-IR spectroscopy, electron microscopy, and rheology. While the tripeptide-based
pure hydrogel was found to be unstable after ca. half an hour, addition of PVA, a water soluble polymer,
increased the temporal and mechanical stability of the hydrogel. A rheological step-strain experiment
demonstrates that the peptide-polymer hydrogel is thixotropic. Results from a fluorescence probe study
and transmission electron microscopy reveal that addition of PVA increases both the fibre diameter and
Received 28th August 2018, entanglement. Circular dichroism spectra of the hydrogels confirm the formation of aggregates with
Accepted 6th December 2018 supramolecular chirality. The thixotropic nature of the hydrogel has been exploited to entrap and release
DOI: 10.1039/c8sm01766b doxorubicin, an anticancer drug, under physiological conditions. Furthermore, an MTT assay of the Fmoc-
tripeptide using AH927 cells confirmed its cytocompatibility, which broadens the utility of the hybrid gel
rsc.li/soft-matter-journal for biomedical applications.

1 Introduction stress, the injected solution in the targeted site will recover its
gel state, allowing it to remain localised in the targeted area.
In recent years, the increasing range of applications of tunable, For this reason, significant efforts have been made to produce
stimuli-responsive and robust gels, especially hydrogels,1–10 hydrogels with the appropriate mechanical properties. How-
of low-molecular-mass gelators (LMMGs) in fields such as ever, in spite of those efforts, there are few reports on the use of
sensors, biomaterials, tissue engineering, and drug delivery thixotropic, injectable hydrogels in the field of biomedical
has attracted the attention of chemists, physicists, materials applications.12–18
scientists, and engineers. However, tenability is frequently found In a few previous reports, it has been established that
at the expense of another important attribute, gel toughness; increasing the water solubility of a peptide facilitates stabilization
many of the gels are observed to be mechanically weak, limiting of its fibrils and generates more stable hydrogel systems.19
their utilization in biomedical applications. Especially for site- In addition, protein/peptide co-assembly has also been reported
specific drug delivery vehicles, hydrogels must be mechano- to help in stabilizing the self-assembled structures of the
responsive (i.e., thixotropic11), such that the gel-to-sol phase hydrogels.20–23 Very recently, Nandi and co-workers have demon-
transitions can be induced isothermally by mechanical pertur- strated that incorporation of poly(ethylene glycol) (PEG) into
bations and the sols reform over time after the cessation of the the hydrogel system of N-fluorenylmethoxycarbonyl-L-tryptophan
mechanical force. In principle, thixotropic hydrogels can be improves the mechanical and thixotropic property of the hybrid
used as hosts for injection in vivo of guest drug molecules using hydrogel.24 In fact, amalgamation of polymers as additives with
a syringe to a targeted site. After the removal of the external supramolecular gels to modify their characteristics has become a
very interesting topic of study.25
a
Department of Chemistry, Indian Institute of Technology Kharagpur, In this approach, a new robust system is obtained by com-
Kharagpur-721 302, India. E-mail: [email protected] bining polymers with an externally switchable supramolecular
b
Department of Chemistry and Institute for Soft Matter Synthesis and Metrology,
gel so that the new system has excellent mechanical stability
Georgetown University, Washington, DC 20057-1227, USA
† Electronic supplementary information (ESI) available: Details of synthesis,
while retaining the characteristics of gels. Polymers can interact
chemical identification, and FT-IR, NMR and HRMS spectra of the tripeptide. with the system either by direct interactions with the LMMG
See DOI: 10.1039/c8sm01766b fibers by the adsorption of the polymers or indirect interactions

This journal is © The Royal Society of Chemistry 2019 Soft Matter, 2019, 15, 433--441 | 433
 View Article Online

Paper Soft Matter

by changing the solution viscosity. In 1999, Hanabusa and

Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

co-workers first reported an increase of gel strength by the

addition of polymers to an LMMG system.26 They observed that
addition of poly(N-vinylpyrrolidone) or PEG to a L-valine-containing
benzene dicarbonyl derivative in 1-propanol enhanced the strength
of the resulting gel. A similar enhancement of gel strength in
cyclohexane was observed after the addition of polystyrene.
However, the reason for these enhancements was unknown.
Since then, a number of other reports have appeared. For
example, Liu and co-workers have demonstrated that the
poly(ethylene-co-vinyl acetate) (PEVA) copolymer can induce a
dramatic change in the physical appearance and nanostructures
of lanosta-8,24-dien-3b-ol:24,25-dihydro-lanosterol (L/DHL) gels
in diisooctyl phthalate (DIOP).27 Before adding PEVA, the L/DHL
system produced opaque gels with needle-like structures,
Fig. 1 The appearance of Fmoc-Val-Car hydrogels in the presence and
whereas a transparent gel with interconnected branched fibres
absence of PVA (1% w/v); inset: chemical structures of Fmoc-Val-Car, DOX
was obtained after PEVA addition. Later, it was shown that the and PVA.
stability of gels of 4-(4 0 -ethoxyphenyl) phenyl-b-O-D-glucoside in
water/1,4-dioxane mixtures was also enhanced by the addition
of poly(2-hydroxyethyl methacrylate) (PHEMA) as a result of the Fmoc-protected tripeptides, obtained by covalently linking an
adsorption of the polymer chains onto the growing fibers.28 Fmoc-amino acid to L-carnosine (Car, a b-alanine-containing
Nandi and co-workers have also shown a similar type of dipeptide), produced mechanically stable hydrogels under
enhancement in mechanical strength in chitosan and folic physiological conditions. However, one of the tripeptides, con-
acid gels.29 taining L-valine (Fmoc-Val-Car, Fig. 1), produced a metastable
Sometimes, however, addition of a polymer has been shown hydrogel which became a sol/solution within half an hour of
to weaken a gel. Adams and co-workers have shown that being formed. Thus, in lieu of the importance of L-carnosine-
addition of dextran to a pH-dependent naphthalene-dipeptide based tripeptide hydrogelators, we report there a study on the
hydrogel altered the viscosity of the solution, which could be effect of a water-soluble polymer on the mechanical strength of
associated with thinner fibers and reduced mechanical strength.30 the hydrogel produced by the Fmoc-protected tripeptide, Fmoc-
The authors attributed the mechanical changes to the increased Val-Car. By adding poly(vinyl alcohol) (PVA, MW = 89 000–98 000),
viscosity of the solution that resulted upon the addition of the we have succeeded in increasing the metastability and mechanical
polymer; the LMMG molecules were less able to diffuse, leading to strength of the Fmoc-Val-Car hydrogel. The hydrogel was char-
decreased mechanical strength. Later, a comparative study on the acterized in detail using various techniques, including fluores-
addition of other polymers such as PEG, poly(vinyl pyrrolidone), cence spectroscopy, microscopy and rheology. An MTT assay of
and poly(acrylic acid) with LMMG was also performed. Surpris- the hydrogelator was also performed to determine its cytocom-
ingly, in all cases, the system exhibited decreased mechanical patibility and we studied the entrapment of an anticancer drug,
strength.31 On the other hand, McNeil and co-workers have shown doxorubicin, DOX (Fig. 1), in the peptide/polymer co-assembly,
that addition of specifically poly(acrylic acid) to a pyridine-based followed by its release kinetics under physiological conditions
organogel leads to the generation of thinner fibers, but with in order to demonstrate the potential use of the stabilized
increased gel strength and lower critical gelation concentration hydrogel in drug delivery applications.
(CGC) values.32 Therefore, the outcome of the interaction of a
polymer with a gelator is not well-understood. It appears that the 2 Experimental section
choice of gelator and the complementary polymer are important
factors in determining whether increased or decreased mecha- 2.1 Materials
nical strength will be observed.24,33,34 L-Carnosine (Car, 99%) and poly(vinyl alcohol) (MW = 89 000–
It is known that a-amino acid-containing peptide molecules 98 000, 99%) were purchased from Sigma-Aldrich (Bangalore,
are susceptible to proteolysis by proteolytic enzymes, a fact that India). Fmoc-L-valine (Fmoc-val, 98%), N-hydroxysuccininimide
limits the utility of peptide-based hydrogel systems as drug (NHS, 98%), and 1,3-dicyclohexylcarbodiimide (DCC, 98%) were
carriers in biomedical applications. For this reason, there is purchased from SRL (Mumbai, India) and were used as received.
a need for the development of proteolytically-stable peptide All organic solvents for purification through column chromato-
hydrogelators. Such stability can be achieved either by intro- graphy were supplied by SRL (Mumbai, India) and were of the
ducing one or more D-amino acid residues or a b-amino acid highest purity (99%). Dichloromethane was dried over CaH2
into an oligopeptide chain. In a recent report, we have demon- (SRL, Mumbai) and methanol was dried using I2/Mg (SRL, Mumbai)
strated an increased proteolytic stability of a number of b-alanine- turnings.35 Milli-Q water (18 MO cm1) was used for the preparation
containing Fmoc-tripeptides that form hydrogels at room of buffers. The AH927 cells were provided by the Department of
temperature at relatively low concentrations.35 Most of these Biotechnology, IIT Kharagpur.

434 | Soft Matter, 2019, 15, 433--441 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Soft Matter Paper

Fmoc-Val-Car was synthesized as reported in our previous of preparation. Each preformed gel was scooped out from a
Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

report and was isolated in the neutral form.35 The chemical wide mouth vial using a spatula to avoid damage to the gel
structure of the Fmoc-tripeptide was identified by Fourier structure and was placed on the rheometer plate. An equili-
transform infrared (FTIR), 1H-NMR (Fig. S1, ESI†), and high- bration time of 30 min was allowed for each sample before
resolution mass spectra (HRMS) (Fig. S2, ESI†). Details can be measurement. Oscillatory stress sweep measurements were
found in the ESI.† carried out at a constant frequency of 1 Hz to obtain storage
moduli (G 0 ) and loss moduli (G00 ).
2.2 Methods and instrumentation The cytotoxicity assay was performed following a standard
The melting point of the tripeptide was measured using an protocol.36 AH927 cells were trypsinized and counted using a
Instind (Kolkata) melting point apparatus with open capillaries. hemocytometer and then seeded in 96-well micro plates at 5 
FT-IR spectra were recorded using a PerkinElmer (Model Spec- 104 cells per ml in DMEM (Dulbecco’s modified Eagle medium)
trum Rx I) spectrophotometer. 1H NMR spectra were recorded complete medium. The cells were incubated at 37 1C in 5% CO2
on an AVANCE DAX-400 (Bruker, Sweden) 600 MHz NMR for 24 h to allow adherence. The medium was replaced with
spectrometer in DMSO-d6 solvent. fresh DMEM in the complete medium and incubated with the
Aqueous buffer (20 mM, pH 7.4) was made by mixing hydrogelators at concentrations of 20, 50, 100, 200 and 500 mg mL1
appropriate volumes of NaH2PO4 and Na2HPO4 solutions. The for 24 h. After incubation, the cells were washed with buffer (pH 7.4)
pH was measured using a digital pH meter (Systronics-335, and 100 mL of MTT solution (1 mg mL1) was added to each well.
Kolkata, India) after standardising the instrument with standard After 3 h of incubation, the MTT solution was removed and replaced
pH 7 and pH 4 buffers. with 200 mL of DMSO. Absorbances were recorded at 570 nm using a
Melting temperatures of the gels were measured by inverting microplate reader (Thermo Scientific Multiskan Spectrum, USA).
the screw-cap vial (average diameter of 10 mm) containing the Each sample was assayed in triplicate. Control samples include cells
gel in a temperature-controlled water bath (JULABO, model F12). with DMSO.
The gel was slowly heated at a rate of 1 1C min1 until the gelated
mass started to flow on tilting of the vial.
The UV-vis spectra of the solutions were recorded using a 3 Results and discussion
UV-2450 UV-VIS spectrophotometer (Japan) in a quartz cuvette
with a path length of 1 cm. Fluorescence spectra measurements 3.1 Gelation behaviour
were performed using a 1 cm2 cuvette on a LS-55 luminescence Gelation studies of the as-synthesized peptide Fmoc-Val-Car
spectrometer (PerkinElmer, U.K.) equipped with a filter polarizer were carried out in 20 mM buffer solutions at different pH
and a thermostated cell holder. Each sample was equilibrated values. The results indicate that the peptide is a good hydro-
for 12 h before recording the spectra. The excitation wavelength gelator at pH 2 and a poor one at pH 7.4. At all other pH values
in all cases was 262 nm and fluorescence was detected at a right examined, the tri-peptide remained insoluble. Although
angle with respect to the excitation beam. enhanced solubility of Fmoc-Val-Car at higher pH values is
The morphology of the hydrogels was determined by trans- expected to produce larger fibrils, strong inter-fibril interaction
mission electron microscopy (TEM), scanning electron micro- can lead to phase separation. This explains why the hydrogel is
scopy (SEM) and atomic force microscopy (AFM) experiments. transparent at pH 2 and translucent at pH 7.4. The detailed
TEM images were taken using a transmission electron micro- gelation properties of Fmoc-Val-Car have been reported in a
scope (FEI-TECNAI G2 20S-TWIN, FEI, USA) operating at an recent paper from this group.35 The hydrogel at pH 7.4 was
accelerating voltage of 120 kV. The hydrogels were drop cast on observed to shrink after 30 min by releasing trapped water, but
carbon-coated copper grids and the samples were air dried. For the gel at pH 2 remained unaffected for several days. This
SEM experiments, the hydrogels were placed on aluminium unusual behaviour was attributed to either loose packing of the
foil and air-dried at room temperature. A layer of gold was peptide molecules or the hydrophobic nature of the peptide
sputtered on top to make a conducting surface, and finally molecules that forced the fibers to phase separate, leaving a
the specimen was transferred into a field emission scanning supernatant liquid. To enhance the metastability of the hydro-
electron microscope (FESEM, Zeiss, Supra-40, Netherland) gel at pH 7.4, the gelation test of Fmoc-Val-Car was carried out
operating at 5–10 kV to get the micrograph. in the presence of different concentrations of a water-soluble
CD spectra were measured on a Jasco J-815 (Japan) spectro polymer, poly(ethylene glycol) (PEG, Mw = 550). However, it
polarimeter using quartz cells of 1 mm path length. Samples failed to stabilize the hydrogel. Thus, we tried another well-
were equilibrated for 12 h before recording the spectra that known biodegradable polymer,37 PVA. Because PVA can interact
were baseline corrected using the appropriate reference solvent. with the gelator molecule through hydrogen-bonds (H-bonds),
Rheological measurements were performed on a Bohlin RS it was expected to enhance the mechanical stability of the
D-100 (Malvern, UK) rheometer using the parallel-plate (PP-20, hydrogel. An initial gelation test was performed in the presence
diameter 20 mm) geometry with a constant tool gap of 100 mm. of 1% PVA (w/v) at pH 7.4. To our satisfaction, the opaque
The rheometer was fitted with a solvent trap and a Peltier hydrogel of the tripeptide transformed into a transparent gel
device that controlled the temperature at 25  0.1 1C. All that did not flow under the force of gravity over a period of even
measurements were performed with matured gels after 10 h one month (see Fig. 1). In the presence of 1% PVA, the solubility

This journal is © The Royal Society of Chemistry 2019 Soft Matter, 2019, 15, 433--441 | 435
 View Article Online

Paper Soft Matter

of Fmoc-Val-Car in the buffer improved, but it required heating the thermal stability of the hybrid hydrogel increases non-
Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

at 60 1C for about 5–10 min to dissolve; in the absence of PVA, linearly with [PVA]; there is strong co-assembly of PVA and
heating at 75–80 1C for about 30 min is required for gelation. peptide molecules.
In order to determine the CGC value and the minimum PVA
concentration required to stabilize the hydrogel, a detailed 3.2 Interaction of Fmoc-Val-Car with PVA
study was carried out. A hydrogel that is stable for more than The self-assembly behaviour of Fmoc-Val-Car with and without
one month could be obtained in the presence of as little as PVA was monitored by measuring the UV-vis spectra and emis-
0.04% (w/v) PVA. Also the CGC value was observed to decrease sion spectra in solution as well as in the gel state (Fig. 3(a)).
from 0.7% (w/v) in pure buffer to 0.5% (w/v) in the presence of UV-vis spectra of Fmoc-Val-Car were measured in DMSO and in
0.04% PVA. Addition of more PVA (4% w/v) led to a further pH 7.4 buffer solutions. Because the molecules remain in the
reduction of the CGC of Fmoc-Val-Car to 0.3% (w/v). These data non-aggregated state in DMSO,24 the absorption peaks in the
are shown in Fig. 2. The plot shows an initial sharp reduction range of 290–297 nm can be ascribed to the p–p* transition of
followed by a slow reduction of the CGC value, indicating the Fmoc moiety. In contrast, when the UV spectra of Fmoc-Val-
very strong co-assembly of the PVA and peptide molecules. Car were measured alone in pH 7.4 aqueous buffer solution,
In the control experiment involving only 4% PVA in pH 7.4 a red shift of the peak from 287 to 301 nm was observed. This
buffer, however, no gelation was achieved. However, thermo- suggests the existence of p–p stacking interactions that facilitate
reversible hydrogel formation by PVA at a very high concentra- the formation of J-aggregates.20 However, after the addition of
tion (15% w/v) induced using a ‘‘repeated freezing-and-melting’’ PVA, a small blue shift of the absorption maximum was noticed.
method has been reported in the literature.38,39 This leads to This indicates the existence of a significant PVA-peptide inter-
the conclusion that a strong interaction between the peptide action which inhibits p–p stacking of the Fmoc groups.
and the polymer reinforces the Fmoc-Val-Car hydrogel network The fluorescence emission spectrum of Fmoc-Val-Car in the
structure. The phenomenon can be linked to the increased solution state (106 M) was also compared with those of pre-gel
fibre thickness and/or density as a consequence of the stronger solutions of Fmoc-Val-Car (3  103 M) with increasing [PVA] as
H-bonding interactions between the polymer and peptide shown in Fig. 3(b). The fluorescence spectrum of Fmoc-Val-Car in
molecules. the solution state reveals a maximum at 312 nm. However, the
Gelation is believed to be a phenomenon similar to crystal- fluorescence spectrum of the Fmoc-Val-Car/PVA gel (0.3% PVA)
lization in which both processes are accompanied by nuclea- not only experiences a red shift to 327 nm, but also a 2.5 increase
tion induced growth.40–43 Since most nucleation processes are in the emission intensity is observed. Furthermore, an increase in
heterogeneous in nature, additives such as dust or other compo- the emission intensity (5.8 times) is observed in the Fmoc-Val-Car/
nents can modify the nucleation rate by including the additional PVA gel when the PVA concentration was increased to 1% (w/v),
nucleating sites. In this way, the gelator may produce higher without any further red shift. The red shift in the emission spectra
numbers of fibres in the gel network. Alternatively, the higher gel of the hydrogels relative to the solution state clearly confirms the
strength may be due to a higher number of nucleating sites and stabilization of the Fmoc groups through extensive p-stacking
higher fibre density. interaction. This type of rise in emission intensity after the
The effect of PVA on the thermal stability of the hybrid inclusion of polymers is very rare, and only a few reports on this
hydrogels was studied by measuring the gel-to-sol transition are available in the literature.24,44 Actually, molecular aggrega-
temperatures (Tgs) of the hydrogels prepared in the presence of tion through p-stacking interaction causes a decrease in inten-
different concentrations of PVA. The results in Fig. 2 show that sity owing to quenching of the excitons by the solvent molecules
after gel formation.45 The rise of fluorescence intensity upon
the addition of PVA can be associated with the increase of solu-
tion viscosity that prevents free rotation of the Fmoc groups,
thereby slowing non-radiative decay processes.

Fig. 3 (a) UV-vis spectra of Fmoc-Val-Car (105 M) in buffer (pH 7.4)

containing 1% PVA (1), buffer (pH 7.4) without PVA (2) and in DMSO (3);
Fig. 2 Variation of CGC and Tgs of Fmoc-Val-Car/PVA hydrogels with (b) fluorescence spectra of Fmoc-Val-Car with increasing PVA concentration
increasing [PVA] at 25 1C; the hydrogels were prepared at their respective in water. Path length of the cell was 1 mm and lex = 265 nm. [Fmoc-Val-Car]
CGC values. for spectrum 1 = 106 M and [Fmoc-Val-Car] for spectra 2 to 6 = 3  103 M.

436 | Soft Matter, 2019, 15, 433--441 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Soft Matter Paper

to ascertain the H-bond formation between Fmoc-Val-Car and

Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

PVA in the hybrid gel state, we have recorded the FTIR spectra
of neat PVA (Fig. S3(b), ESI†). The spectrum shows the presence
of a peak at 3424 cm1 corresponding to O–H vibrations and is
shifted to 3435 cm1 in the hybrid gel state. Shifting of the O–H
vibrations toward higher frequency clearly confirms H-bonding
interactions with Fmoc-Val-Car, but not with PVA molecules.
Thus, the sum of the spectral changes clearly suggests that
Fig. 4 Fluorescence spectra of thioflavin T (ThT,6  105 M) in pH 7.4
intermolecular H-bond formation is one of the principal driving
buffer as a function of increasing [PVA]: (a) in the presence of Fmoc-Val-Car
(105 M) (1 = 0 mg, 2 = 0.1 mg, 3 = 0.3 mg, 4 = 0.5 mg, and 5 = 0.7 mg PVA
forces for the co-assembly of PVA and Fmoc-Val-Car.
in each 2 mL solution); (b) in the absence of Fmoc-Val-Car but with PVA Based on the above results, a model for the incorporation of
(1 = 0.1 mg, 2 = 0.5 mg, and 3 = 0.7 mg PVA in each 2 mL solution). The lex the two components into the hydrogel network is proposed as
and path length of the cell were 440 nm and 1 cm, respectively. shown in Fig. 5. It is based on the aforementioned observations
(1) from FTIR spectra that H-bonding interactions develop
during gel formation, possibly involving the amide, urethane,
To study the effect of PVA on the network structure of the –COOH and imidazole groups and (2) from UV-vis absorption
tripeptide hydrogel, fluorescence measurements (Fig. 4(a)) using and fluorescence spectra that there are p-stacking interactions
thioflavin T (ThT) as a probe were performed at increasing PVA between the Fmoc groups. In this regard, it is important to
concentrations; this approach is a commonly used method to remember that the imidazole ring and the –COOH groups in
examine whether fibril formation is occurring during self- the Fmoc-Val-Car hydrogelator are deprotonated at pH 7.4; the
assembly:46–48 an enhancement in ThT fluorescence is an pKa value of the former is B6.8352 and that of the latter is 2.64.52
indicator of fibril formation by amphiphilic molecules in There should be competition for H-bond formation: (i) the
solution.49–51 The experiments were performed at a fixed con- gelator molecules may form intermolecular H-bonds among
centration of Fmoc-Val-Car (105 M) and ThT (6  105 M) and themselves, leaving PVA intact; (ii) the imidazole and –COO
increasing concentrations of PVA. The fluorescence spectra groups may form H-bonds with water molecules; and (iii) PVA
were recorded by exciting the resultant solutions at 440 nm can interact with the gelator molecules laterally through
after equilibrating them for 24 h. The spectra in Fig. 4 show a H-bonds with the imidazole or –COO groups. Because TEM
gradual increase of emission intensity with increasing PVA images (discussed below) suggest that the fibre diameters and
concentration, suggesting more (or a higher weight fraction density increase after the addition of PVA, we conclude that PVA
of) fibrils upon increasing PVA concentration. Thus, the results interacts with the peptide backbone laterally, forming H-bonds
as shown in Fig. 4 provide indirect proof that PVA induces with the imidazole and –COO groups. The next layer is added
formation of fibrils which produce a 3-dimensional (3-D) gel in such a way that the somewhat disordered alkyl chains of PVA
network that entraps increasing amounts of water. Results from create the crystallites observed in the TEM images. As a result,
a control experiment in which ThT fluorescence was measured lateral growth of the fibres occurs through H-bond forma-
in the presence of only PVA, but without any Fmoc-Val-Car tion with PVA. The longitudinal growth of the fibres, on the
(Fig. 4(b)) are consistent with this conclusion; the intensity of
ThT fluorescence increases, but to a much smaller extent than
when Fmoc-Val-Car is present. The peptide molecules dissolved
very easily in the buffer solution, in the presence of PVA. Thus,
it appears that PVA interacts with the Fmoc-Val-Car molecules
through the formation of additional H-bonds involving its –OH
groups. To envisage the participation of all functional groups in
the hydrogel formation, FT-IR spectra of Fmoc-Val-Car as a neat
solid and as an Fmoc-Val-Car/PVA xerogel were recorded
(Fig. S3, ESI†). The observed stretching vibrations at 3418 and
3293 cm1 corresponding to the O–H and N–H vibrations,
respectively, in the solid Fmoc-Val-Car are merged, producing
a broad peak located at 3435 cm1 that suggests H-bond
formation involving all the functional groups during hydrogela-
tion. The CQO stretching vibration of the –COOH group and the
amide groups of solid Fmoc-Val-Car were observed at 1683 and
1654 cm1, respectively. After hydrogel formation, these peaks
again merged and shifted to 1649 cm1 with extensive broad-
ening. In the solid sample, a sharp peak of the N–H bending
vibration at 1536 cm1 suffered a reduction in intensity and Fig. 5 Schematic model for Fmoc-Val-Car and PVA molecules in the
was not distinguishable after hydrogel formation. Moreover, hydrogel.

This journal is © The Royal Society of Chemistry 2019 Soft Matter, 2019, 15, 433--441 | 437
 View Article Online

Paper Soft Matter

other hand, occurs through the p-stacking interactions among 104 M concentration of the gelator. Consequently, the spec-
Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

the Fmoc groups. trum does not exhibit any discernible signals; the Fmoc groups
of molecularly dispersed Fmoc-Val-Car must be able to rotate
3.3 Gel morphology freely. By contrast, the Fmoc-Val-Car/1% PVA hybrid gel dis-
The TEM images of the dried Fmoc-Val-Car hydrogels with and played a number of signals, confirming the formation of a more
without PVA were recorded to distinguish the differences in gel ordered structure in which rotation of the Fmoc group is
morphology caused by the presence of the polymer (Fig. 6). In restricted. The positive peak centred at 230 nm may be indicative
the absence of PVA, the Fmoc-Val-Car gel exhibits fibres with of superhelical arrangements of the peptide molecules in the
average diameters of 6  3 nm. In the presence of PVA, the self-assemblies.54 A number of positive peaks with shoulders
fibres tend to entangle to a great extent and there is an increase between 270 and 305 nm are associated with the p–p* transitions
in fibre diameter that depends on PVA concentration. Also, of the Fmoc groups that are stacked face-to-face in the helical
large PVA crystallites, which act as junction points for fibre aggregates (Fig. S4(a), ESI†).55,56 Even one day after gel prepara-
growth, form enhanced gelation. The crystallites are formed by tion, the CD spectrum of the polymer-peptide co-assembly does
the physical cross-linking of the long amorphous chains of PVA.38 not exhibit any significant changes. The CD spectrum of the
At 0.04% PVA, the fibre diameter was 15  5 nm, whereas at 1% hydrogel in the absence of PVA, recorded immediately after its
PVA, it increased to 27  5 nm. This increase was accompanied preparation, shows only one sharp, red-shifted positive peak
by increased fibre branching (thereby increasing fibre density). centred at 305 nm that is indicative of stacking of Fmoc groups.
These electron micrographs demonstrate that PVA plays a key However, this band becomes less prominent with time in a
role in the aggregation processes. Although not very common, fashion that is correlated with the loss of gel stability.
this type of increase in fibre diameter after the addition of
polymers has been reported by others.33 Most commonly, a 3.4 Thixotropic behavior
decrease in fibre diameter is found upon the addition of a The viscoelastic properties of the hydrogels with varying PVA
polymer.24,29,53 However, the possibility that the change in the concentrations were determined using oscillatory amplitude
network morphology is due to the removal of water during the and frequency sweep measurements performed at their respec-
sample preparation for the TEM cannot be ruled out. tive CGC values (Fig. 7). These data allow the yield stress (sy)
The chiral nature of the Fmoc-Val-Car tripeptide leads to the values of different gels and the impact of PVA on the gel net-
expectation of chiral, self-assembled supramolecular struc- work structure to be compared. The frequency sweep measure-
tures. Unfortunately, insufficient resolution and the effects of ments show that the storage modulus (G 0 ) is higher than the
drying the samples may have masked any twisting of the fibers. loss modulus (G00 ) and both remain almost independent of
Therefore, circular dichroism (CD) spectra of the hydrogels
were recorded to explore the existence of self-assembled struc-
tures with supramolecular chirality (Fig. S4, ESI†). CD spectra
of Fmoc-Val-Car in solution as well as in the hybrid gel con-
taining 1% PVA are reported. In our previous work,35 it has
been shown that no self-assembly occurs in solution at the

Fig. 7 G 0 and G00 as a function of shear stress (a–c; frequency = 1 Hz) and
frequency (d–f; shear stress = 0.1 Pa) for Fmoc-Val-Car hydrogels at
Fig. 6 TEM images of Fmoc-Val-Car (initially 104 M) solution in the pH 7.4 in the presence of (a, d) 0.1%, (b, e) 0.5%, and (c, f) 1% PVA. The
presence of (A) 0% (w/v), (B) 0.04%, (C) 0.5%, and (D) 1% PVA. hydrogels were prepared at their respective CGC values.

438 | Soft Matter, 2019, 15, 433--441 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Soft Matter Paper

Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

Fig. 9 Release of DOX from an Fmoc-Val-Car/1% PVA hydrogel at pH 7.4:

(a) images of the vials during different experimental steps; (b) release
profile of DOX from an Fmoc-Val-Car/1%PVA hydrogel at pH 7.4. The
concentration of DOX is 0.1 mg mL1 and the error bars are calculated
using data from two experiments.

Fig. 8 Continuous measurements at alternating 0.001% and 1% strain

(frequency = 1 Hz) for the hydrogel of Fmoc-Val-Car/PVA (1%) at pH 7.4
that had been equilibrated for 12 h before measurements were taken. The
hydrogel was prepared at its CGC value.

frequency throughout the range examined; these are hallmarks

of a true gel. The sy values of the gels were determined in the
presence of 0.1, 0.5 and 1% PVA. The increasing order of sy
values with the increase of [PVA] implies that the hydrogels
become mechanically stronger as [PVA] is increased: 10 Pa at
0.1% PVA and 38 Pa at 1% PVA.
The data in Fig. 8 show that the addition of PVA makes the
Fig. 10 Cell survival study (by MTT assay) of the hydrogelator Fmoc-Val-
hydrogel thixotropic. Initially, the measurements were made at Car against the AH 927 cell line.
a minimum strain of 0.001% (i.e., where G 0 4 G00 ). Then, the gel
was converted to a viscous-liquid at 1% strain (i.e., where
G00 4 G 0 ). Thereafter, the initial two steps were repeated, leading The release profile thus obtained is presented in Fig. 9(b). The
to small decreases in gel recovery in subsequent cycles. Almost plot indicates that the initial rate of drug release is slow during
90% of the gel viscoelasticity was recovered after the first the first 10 hours, and increases thereafter with a total release
disruption. Based on other studies,57 it is reasonable to assume of 80% of the drug after 72 hours. Thereafter, the hydrogel
that the H-bonding (and other) interactions are being disrupted became weak and ruptured. The process of drug release initially
here when the gels are destroyed mechanically and they are starts from the surface of the hydrogel by the dissolution of the
re-established over time when the strain is very small. water-soluble drugs. Then the rest of the drug molecules are
released through the diffusion process which is maintained by
3.5 Drug encapsulation and release study the penetration of the water molecules into the gel network.58
The thixotropic nature of the hydrogel is an important property From the data in Fig. 9 it appears that the diffusion of water
for an injectable drug delivery system. It allows a drug to be into the gel matrix is a rate-controlling factor.
encapsulated into a hydrogel matrix isothermally (i.e., avoiding
3.6 Cytotoxicity study
heating and cooling protocols that may affect the drug’s efficacy).
Also, the drug-encapsulated hydrogel can be converted into a An MTT cell viability assay of this hydrogelator Fmoc-Val-Car
free-flowing solution by mechanical force and then injected using against AH 927 cells (non-cancerous cells) was performed. The
a syringe into the spot where the drug is needed and then the hydrogelator at five different concentrations was incubated
hydrogel will be reconstituted in vivo over time. As an example, with AH 927 cells for 24 h at 37 1C (Fig. 10). The data suggest
the anticancer drug, DOX, was encapsulated at 0.1 mg mL1 into that all the samples were highly cell compatible; almost 90–95%
the hybrid hydrogel matrix utilizing its thixotropic property. cells remained alive at the highest concentration of Fmoc-Val-Car,
Thus, initially 1 mL of a Fmoc-Val-Car/PVA (1%) gel at pH 7.4 500 mM, used. Therefore, it can be concluded that the hydro-
was prepared and DOX was encapsulated in it by adding DOX- gelator Fmoc-Val-Car and its analogues can be used in biomedical
dissolved buffer on the top of the gel followed by shaking the gel applications.
and equilibration of the gel for 12 h to make a homogeneous
dispersion of the drug. Then, 1 mL of buffer (pH 7.4) was care- 4 Conclusions
fully placed on top of the gel surface (Fig. 9(a)). The release of the
drug from the gel matrix to the buffer solution through slow The poor mechanical stability of many hydrogels limits their
diffusion was monitored via UV-vis spectroscopy by measuring use as functional materials. In this work, a PVA embedded
the absorbance values at 490 nm at different time intervals. hybrid hydrogel of a tripeptide, Fmoc-Val-Car, was produced in

This journal is © The Royal Society of Chemistry 2019 Soft Matter, 2019, 15, 433--441 | 439
 View Article Online

Paper Soft Matter

pH 7.4 buffer and at room temperature. The tripeptide-based 12 A. Pasc, P. Gizzi, N. Dupuy, S. Parant, J. Ghanbaja and
Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

hydrogel is not stable for more than half an hour in the absence C. Gérardin, Tetrahedron Lett., 2009, 50, 6183–6186.
of added PVA. The room-temperature stability of the hydrogel 13 H. Liu, Y. Hu, H. Wang, J. Wang, D. Kong, L. Wang, L. Chen
as well as its mechanical stability increased significantly on and Z. Yang, Soft Matter, 2011, 7, 5430–5436.
the addition of PVA. Structurally, PVA increases both the fibre 14 S. Roy, A. Baral and A. Banerjee, Chem. – Eur. J., 2013, 19,
entanglement and fibre diameter. A fluorescence study indicates 14950–14957.
that the increase of room-temperature stability of the hydrogel is 15 Y. Wang, R. Huang, W. Qi, Z. Wu, R. Su and H. Zhimin,
associated with the higher density of fibres generated. In addition, Nanotechnology, 2013, 24, 465603.
the addition of PVA imparts a thixotropic behaviour to the peptide 16 C. Yan, A. Altunbas, T. Yucel, R. P. Nagarkar, J. P. Schneider
hydrogels. The thixotropic nature was utilized to effect the encap- and D. J. Pochan, Soft Matter, 2010, 6, 5143–5156.
sulation and slow release of an anticancer drug, doxorubicin, at 17 L. Yu and J. Ding, Chem. Soc. Rev., 2008, 37, 1473–1481.
physiological pH. An 80% release of doxorubicin from the gel 18 M. Guvendiren, H. D. Lu and J. A. Burdick, Soft Matter, 2012,
matrix was achieved after 72 hours. Furthermore, the cytocompat- 8, 260–272.
ibility of the tripeptide Fmoc-Val-Car against AH927 cells broadens 19 D. M. Ryan, T. M. Doran and B. L. Nilsson, Chem. Commun.,
the utility of the hydrogel in biomedical applications. 2011, 47, 475–477.
20 Y. Kuang, D. Yuan, Y. Zhang, A. Kao, X. Du and B. Xu,
RSC Adv., 2013, 3, 7704–7707.
Conflicts of interest 21 D. I. Rożkiewicz, B. D. Myers and S. I. Stupp, Angew. Chem.,
There are no conflicts to declare. Int. Ed., 2011, 50, 6324–6327.
22 D. M. Ryan, T. M. Doran and B. L. Nilsson, Chem. Commun.,
2011, 47, 475–477.
Acknowledgements 23 Y. M. Abul-Haija, S. Roy, P. W. J. M. Frederix, N. Javid,
V. Jayawarna and R. V. Ulijn, Small, 2014, 10, 973–979.
We thank the Indian Institute of Technology Kharagpur for the 24 P. Chakraborty, S. Mondal, S. Khara, P. Bairi and A. K. Nandi,
partial support of this work. R. D. M. thanks CSIR, New Delhi J. Phys. Chem. B, 2015, 119, 5933–5944.
for a research fellowship (09/081(1157)/2012-EMR-I). The authors 25 D. J. Cornwell and D. K. Smith, Mater. Horiz., 2015, 2,
are thankful to Dr Kiran Patruni, Department of Agricultural 279–293.
Science, IIT Kharagpur for his assistance with the rheological 26 K. Hanabusa, A. Itoh, M. Kimura and H. Shirai, Chem. Lett.,
measurements. We also thank the Department of Biotechnology, 1999, 767–768.
IIT Kharagpur for providing AH927 cells and for assistance with 27 X. Y. Liuand and P. D. Sawant, Angew. Chem., Int. Ed., 2002,
the biological measurements. R. G. W. thanks the US National 41, 3641–3645.
Science Foundation for its support through Grant CHE—1502856. 28 M. M. Smith and D. K. Smith, Soft Matter, 2011, 7,
4856–4860.
Notes and references 29 P. Chakraborty, B. Roy, P. Bairi and A. K. Nandi, J. Mater.
Chem., 2012, 22, 20291–20298.
1 A. Friggeri, B. L. Feringa and J. van Esch, J. Controlled 30 L. Chen, S. Revel, K. Morris, D. G. Spiller, L. C. Serpell and
Release, 2004, 97, 241–248. D. J. Adams, Chem. Commun., 2010, 46, 6738–6740.
2 J. C. Tiller, Angew. Chem., Int. Ed., 2003, 42, 3072–3075. 31 G. Pont, L. Chen, D. G. Spiller and D. J. Adams, Soft Matter,
3 D.-L. Liu, X. Chang and C.-M. Dong, Chem. Commun., 2013, 2012, 8, 7797–7802.
49, 1229–1231. 32 Y. J. Adhia, T. H. Schloemer, M. T. Perez and A. J. McNeil,
4 J. Chen and A. J. McNeil, J. Am. Chem. Soc., 2008, 130, Soft Matter, 2012, 8, 430–434.
16496–16497. 33 P. Chakraborty, S. Das, S. Mondal, P. Bairi and A. K. Nandi,
5 X. Wang, A. Horii and S. Zhang, Soft Matter, 2008, 4, Langmuir, 2016, 32, 1871–1880.
2388–2395. 34 J. R. Moffat and D. K. Smith, Chem. Commun., 2008,
6 A. R. Hirst, B. Escuder, J. F. Miravet and D. K. Smith, Angew. 2248–2250.
Chem., Int. Ed., 2008, 47, 8002–8018. 35 R. Das Mahapatra, J. Dey and R. G. Weiss, Langmuir, 2017,
7 S. Banerjee, R. K. Das and U. Maitra, J. Mater. Chem., 2009, 33, 12989–12999.
19, 6649–6687. 36 P. Laskar, B. Saha, S. K. Ghosh and J. Dey, RSC Adv., 2015,
8 A. Dawn, T. Shiraki, S. Haraguchi, S.-i. Tamaru and 5, 16265.
S. Shinkai, Chem. – Asian J., 2011, 6, 266–282. 37 A. Takasu, K. Aoki, M. Tsucyia and M. Okada, J. Appl. Polym.
9 X. Du, J. Zhou, J. Shi and B. Xu, Chem. Rev., 2015, 115, Sci., 1999, 73, 1171–1179.
13165–13307. 38 E. Yokoyama, I. Masada, K. Shimamura, T. Ikawa and
10 D. B. Amabilino, D. K. Smith and J. W. Steed, Chem. Soc. K. Monobe, Colloid Polym. Sci., 1986, 264, 595–601.
Rev., 2017, 46, 2404–2420. 39 S. R. Stauffer and N. A. Peppas, Polymer, 1992, 33, 3932–3936.
11 V. A. Mallia and R. G. Weiss, Soft Matter, 2016, 12, 40 M. A. Rogers and A. G. Marangoni, Langmuir, 2009, 25,
3665–3676. 8556–8566.

440 | Soft Matter, 2019, 15, 433--441 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Soft Matter Paper

41 X. Huang, S. R. Raghavan, P. Terech and R. G. Weiss, J. Am. 50 R. Orbach, L. Adler-Abramovich, S. Zigerson, I. Mironi-Harpaz,
Published on 07 12 2018. Downloaded by Centro de Investigacion y Desarrollo Tecnologico en Electroquimica SC on 2020/02/13 23:08:26.

Chem. Soc., 2006, 128, 15341–15352. D. Seliktar and E. Gazit, Biomacromolecules, 2009, 10, 2646–2651.
42 G. Tan, V. T. Johnand and G. L. McPherson, Langmuir, 2006, 51 V. Castelletto, G. Cheng, B. W. Greenland, I. W. Hamley and
22, 7416–7420. P. J. F. Harris, Langmuir, 2011, 27, 2980–2988.
43 X. Huang, P. Terech, S. R. Raghavan and R. G. Weiss, J. Am. 52 Carnosine. http://www.druglead.com/cds/carnosine.html.
Chem. Soc., 2005, 127, 4336–4344. 53 Y. J. Adhia, T. H. Schloemer, M. T. Perez and A. J. McNeil,
44 P. Chakraborty, S. Das, S. Mondal, P. Bairi and A. K. Nandi, Soft Matter, 2012, 8, 430–434.
Langmuir, 2016, 32, 1871–1880. 54 Z. Yang, H. Gu, Y. Zhang, L. Wang and B. Xu, Chem.
45 S. Pan, S. Luo, S. Li, Y. Lai, Y. Geng, B. He and Z. Gu, Chem. Commun., 2004, 208–209.
Commun., 2013, 49, 8045–8047. 55 D. M. Ryan, T. M. Doran, S. B. Anderson and B. L. Nilsson,
46 S. Fleming, S. Debnath, P. W. J. M. Frederix, T. Tuttle and Langmuir, 2011, 27, 4029–4039.
R. V. Ulijn, Chem. Commun., 2013, 49, 10587–10589. 56 D. M. Ryan, S. B. Anderson, F. T. Senguen, R. E. Youngman
47 R. Vegners, I. Shestakova, I. Kalvinsh, R. M. Ezzell and and B. L. Nilsson, Soft Matter, 2010, 6, 475–479.
P. A. Janmey, J. Pept. Sci., 1995, 1, 371–378. 57 Y. Zhang and R. G. Weiss, J. Colloid Interface Sci., 2017, 486,
48 A. K. Das, R. Collins and R. V. Ulijn, Small, 2008, 4, 279–287. 359–371.
49 A. Mahler, M. Reches, M. Rechter, S. Cohen and E. Gazit, 58 D. Das, P. Ghosh, S. Dhara, A. B. Panda and S. Pal, ACS Appl.
Adv. Mater., 2006, 18, 1365–1370. Mater. Interfaces, 2015, 7, 4791–4803.

This journal is © The Royal Society of Chemistry 2019 Soft Matter, 2019, 15, 433--441 | 441