Drug release profile and reduction in the in vitro burst release from pectin/HEMA hydrogel nanocomposites crosslinked with titania

Abstract

This work describes the drug release profile and the initial burst release from covalent hydrogel nanocomposites composed of pectin, hydroxyethyl methacrylate (HEMA) and titania (TiO₂). Vitamin B12 (Vit-B₁₂), a highly water-soluble substance, was used as a model drug. We studied the water transport profiles over a wide pH range, the moduli of elasticity (E), the morphological properties and the Vit-B₁₂ release kinetics from these hydrogels. The initial release burst was reduced by crosslinking titania with vinylated pectin and HEMA. A reduction of up to ca. 60% was observed when compared with pure pectin/HEMA hydrogel. To gain insight into the burst release phenomenon, the experimental data were adjusted to diffusive-based models that include a rate constant of release (k). A decrease in the values of k was related to a reduction in the burst effect. The release mechanism of Vit-B₁₂ from the pure hydrogels was governed by both Fickian diffusion and macromolecular relaxation, which are the driving forces for release. Upon addition of titania, the contribution of macromolecular relaxation to the release was minimized, suggesting a tendency towards Fickian diffusion. Furthermore, titania played a significant role in improving mechanical properties. Hydrogel nanocomposites showed a marked increase in E compared with pure hydrogels. This increase was found to be the result of an apparent increment in the cross-linking density, owing to chemical bonds of titania with the hydrogel. The proposed materials were demonstrated to be biocompatible with cells, showing good pharmacological potential.

---

Introduction

Hydrogels are a class of soft materials formed by chemically and/or physically cross-linked polymers, which are somewhat responsible for their transient nature. They are sufficiently versatile and absorb a large amount of water and/or biological fluids without their three-dimensional form being dissolved into the solvent.^1,2 Soft smart materials have been extensively studied for the production of polymer-based controllable systems, which are attractive in the controlled release of drugs.^3–6 Interest in such materials has increased considerably over the last decades owing to their high hydrophilicity, nontoxicity and biodegradability.^7,8 However, the hydrogels show some persistent problems related to their three-dimensional structure that have often limited their applications as soft biomaterial devices, such as the following: (i) poor mechanical performance, owing both to a random arrangement of cross-linkers and to a wide distribution of polymer lengths between crosslinks,⁹ and (ii) high initial drug release rate, a phenomenon typically known as burst release, which can lead to toxic levels.¹⁰ These limitations are further compounded when polysaccharides or related natural polymers are used to prepare hydrogels. On the other hand, hydrogels incorporating polysaccharides are of particular importance to biomaterial devices owing to their nontoxicity, biodegradability, availability, and efficiency of application.¹¹ The idea is to impart these properties into the hydrogel, which can be achieved by chemical approaches such as radical polymerization by complementary groups, ionic interactions by complexation among polyelectrolytes, crystallization, and so on.¹²

The burst effect can be very interesting for applications in which high drug release rate is important at early times, such as in dermal uses.¹³ This is an example of a particular case in which the typical release rate can match the needs of a specific application. In the majority of cases, however, such an effect is undesirable because it is unpredictable and cannot be controlled, leading to an increasing concentration of the drug in the blood flow that can go beyond the therapeutic level. This affects the efficiency of the device from which drug release into the physiological environment is expected to be sustained for a relatively extensive time period.¹⁴ Many efforts have been made to create approaches that allow main variables to be changed to reduce or eliminate the initial drug burst release, providing a soft, smart material system with adequately improved release characteristics.^13,15–17

In order to solve problems associated with burst release and the mechanical properties of soft materials, inorganic nanoparticles appear suitable for both regulating the burst effect by creating an additional barrier against diffusion and meeting the stiffness needs.

The interactions between the inorganic and the organic components of a hydrogel nanocomposite allow the formation of molecular links,¹⁸ providing an excellent dispersion of particles throughout the device. The combination of inorganic nanoparticles with one or more polymers at a molecular level provides enhanced hydrogel properties, such as high hardness, elastomeric properties and control of drug release, as a result of molecular interaction of surface charges with the polymer.¹⁹ Hydroxyapatite,²⁰ silica,²¹ and titanium dioxide²² are typical examples of inorganic compounds that have been used to enhance the physical and biological properties of polymers.

In light of these concepts, titania nanoparticles (TiO₂) can provide an efficient and convenient approach to obtaining improved mechanical properties and controlled initial burst release. Owing to its high hardness, low chemical reactivity and toxicity, and high index refraction,²³ titania is of particular interest in developing biomaterial devices for drug delivery. To prepare such devices, two encouraging approaches can be highlighted. One is to use titania nanoparticles as a chemical cross-linker by cross-linking surface-functionalized nanoparticles with polymers, and the other is to use them as a nanofiller by mechanically trapping them among the polymer networks. The first approach comes closest to providing materials stable enough to ensure that titania does not diffuse from the swollen hydrogel.

This work aims at preparing polysaccharide-based hydrogel nanocomposites for drug delivery with reduced initial burst release rate. This study employed pectin and hydroxyethyl methacrylate (HEMA) as the polymers and titania as the chemical cross-linker. Pectin was chosen owing to its biodegradability and controllable biologic activity, while HEMA readily experiences hydrogelation. A chemical modification approach was used to convert titania from a material with low chemical reactivity to a covalent cross-linker for further radical cross-linking reaction with polymers. In the hydrogel, titania is expected to play a key role in reducing the initial burst release and improving the physical-chemistry properties. To investigate burst release from the hydrogels, vitamin B12 (Vit-B₁₂), a highly water-soluble substance, was used as a model drug.

Experimental section

Materials

Pectin from citrus peel ≥74.0% (galacturonic acid, Sigma-Aldrich), glycidyl methacrylate 97% (GMA, Aldrich), absolute ethanol P.A. 99.5% (Nuclear, Brazil), hydrochloric acid 37% (HCl, Fmaia-Brazil), hydroxyethyl methacrylate 97% (HEMA, Aldrich), sodium persulfate ≥98%, (Aldrich), and vitamin B12 ≥98.5% (Aldrich) were obtained from commercial sources and used as received. The titania nanoparticles were produced using scCO₂-assisted expansion of sol–gel precursor. This approach has been described and characterized in a previous work.²⁴

Preparation of vinylpectin

Twelve grams of pectin were added to 480 mL of distilled deionized water while stirring. After homogenization, hydrochloric acid was added dropwise to the solution until pH 3.5 was reached, and then 1.29 mL GMA was added. The mixture was kept under stirring for 24 h at 60 °C. The resultant product was precipitated in ethanol and centrifuged at 7000 rpm (Sorvallenged XT/XTR) at 10 °C three times to remove residues and impurities. The vinylpectin was freeze-dried for 24 h.

Preparation of vinyltitania

One hundred milliliters of distilled deionized water at 60 °C was added to 0.1 g of titania while stirring. Concentrated hydrochloric acid was added dropwise until pH 3.5 was obtained. Subsequently, 330 μL of GMA was added to the suspension, which was stirred for 12 h at 1000 rpm. The product was washed several times with ethanol to remove residues and impurities, centrifuged at 7500 rpm at 10 °C, and freeze-dried for 24 h.

Preparation of pure hydrogels and nanocomposites

Ten milliliters of distilled deionized water at room temperature were added to predetermined amounts of vinylpectin, vinyltitania, and HEMA. Table 1 summarizes the amounts of the reactants. The hydrogel-forming suspensions were sonicated with the use of an ultrasonic oscillation probe (Cole-Parmer® 500, model EW-04711-40) at a frequency of 20 kHz for 65 s. For hydrogelation, 23 mg of sodium persulfate was added to the stirred suspension, which was then sonicated for 60 s. Under ultrasound, the temperature is raised to ca. 60 °C as a result of the collapse of bubbles, which provides intense local heating.²⁵ In such a condition, the sodium persulfate undergoes homolytic cleavage, producing two sulfate radical ions. Then, the radical ions attack the vinyl groups from HEMA, vinylpectin and vinyltitania, thus forming the hydrogel. The obtained hydrogel nanocomposites were washed in Milli-Q water for 24 h to be characterized. The water was renewed at intervals of 8 h. The samples, swollen to equilibrium, were freeze-dried to preserve their porous morphology, which is the polymer arrangement naturally organized upon hydrogel swelling. The samples without titania were labelled as follows: (30-70), (70-30) and (50-50). The terms designated in parentheses refer to pectin and HEMA (% w/w), respectively. For hydrogels with titania, the nanocomposites, a third term indicating the amount of nanoparticles (% w/w) with respect to total weight of polymers was included, such as (30-70-0.5), (70-30-0.5) and (50-50-0.5).

Table 1 Compositions of pectin, HEMA and titania used in the hydrogel-forming mixtures

Composition of hydrogels
Sample	Pectin^a (%)	HEMA^b (%)	TiO2–M^c (%)
a Pectin (% w/w).b HEMA (% w/w).c Amount of nanoparticles (% w/w) with respect to total weight of polymers.
1	30	70	0
2	70	30	0
3	50	50	0
4	30	70	0.5
5	70	30	0.5
6	50	50	0.5

---

Swelling kinetics

The swelling degree (SD) was defined and determined as the ratio of the weight of water in the swollen hydrogel to the weight of the dry hydrogel (eqn (1)). The dry samples were brought to aqueous solutions of pH 2, 7 and 10 at 37 °C and left to swell until equilibrium. In order to track the evolution of swelling, the hydrogels were removed from the solutions at predetermined periods, wiped off carefully to remove excess water droplets on the surface, and weighed until no weight variation was observed.where W_s is the weight of the hydrogel swollen to equilibrium, and W_d is the weight of dry gel.

Mechanical properties

Mechanical tests were performed by compressing the hydrogels to 1 mm deformation using a texture analyzer-TAX.T2i equipped with a 5 kg load cell. A 0.5 mm diameter circular probe was adjusted to go down onto the hydrogel surface moving at 2 mm s⁻¹. The tests were performed in four replicates using swollen hydrogels with a 10 cm² surface. The samples were swollen to equilibrium in an aqueous solution of pH 7 at 37 °C. Each measurement was performed in less than one minute to prevent loss of water. The data generated by the equipment were force and displacement, which were converted to stress (σ) and deformation (λ) according to eqn (2).²⁶where σ is compression stress, F is the applied force, A is sectional area of the probe that compresses the hydrogel to a 1 mm depth, and λ is apparent deformation. The compression elastic modulus (E) was determined from the slope of the linear dependence of eqn (2).

The apparent cross-linked density, ν_e, was determined using eqn (3).

where ϕ_p,0 and ϕ_p are the polymer volume fractions in the relaxed state (shortly after the hydrogelation) and in the swollen state, respectively. R is the gas constant, and T is temperature given in Kelvin (K).

Cytotoxicity assay

VERO cells grown in DMEM plus 10% fetal bovine serum (FBS) and 50 μg mL⁻¹ gentamicin were distributed in a 96-well microplate at 2.5 × 10⁵ cells per well and incubated in a humid atmosphere with 5% CO₂ at 37 °C until a confluent monolayer was formed. The medium was then removed, and 100 μL of the suspension of powdered hydrogels in DMEM (1 mg mL⁻¹) was added in triplicate. A control that used cells without the addition of the solutions was also included. The plate was incubated again in a humid chamber at 37 °C with 5% CO₂ for 72 h. Viable cells were detected using the sulforhodamine B colorimetric method.²⁷ After culture, the monolayers were washed with a saline solution of phosphate buffer at pH 7.4 and fixed using 50 μL of 10% trichloroacetic acid solution at 4 °C for 1 h. After that, the cells were washed with running water and dried at room temperature. Fifty microliters of sulforhodamine B were added to all of the wells, and after a 30 min incubation period at 37 °C, the plates were washed three times with 1% acetic acid solution, and then 150 μL of 10 mM Tris base was added to each well. Plates were stirred, and the optical densities (OD) were read at 530 nm in an ELISA reader (Bio-Tek FL-600 Microplate Fluorescence Reader), and the cytotoxic concentration for 50% of VERO cells (CC₅₀) was determined through linear regression.

Vit-B₁₂ loading onto and its release from the hydrogels and nanocomposites

Vit-B₁₂ was added to the hydrogel-forming emulsion to be loaded during the hydrogelation. The amount of drug loaded onto hydrogels corresponded to 10% wt with respect to the weight of reactants used for hydrogelation. The as-prepared hydrogels loaded with Vit-B₁₂ were left to dry in a ventilated oven at 37 °C until constant weight was observed. The as-prepared hydrogels were brought into a glass reactor containing 50 mL of buffer solution of pH 2, 7 or 10 while stirring. Aliquots of 5 mL were collected at specified periods, and then absorption readings were made at 360 nm in a UV-vis spectrophotometer (Shimadzu, UV mini 1240). These aliquots were brought back into the reactor after the readings to prevent variations in volume.

Characterizations

FTIR spectra of titania were recorded in a Bomen FT-IR model MB100 spectrophotometer operating at the spectral region from 4000 to 400 cm⁻¹. Spectra were obtained using sample KBr pellets, after being carefully blended. A total of 128 scans were run to achieve a resolution of 4 cm⁻¹.

¹H NMR and ¹³C-CP/MAS NMR spectra were recorded on a Varian spectrometer (model Mercury Plus BB) by applying frequencies of 300.059 MHz and 74.47 MHz for ¹H and ¹³C nuclei, respectively. To record the ¹H NMR spectra, 20 mg of the samples were dissolved in 0.7 mL of D₂O containing 0.05% 3-(trimethylsilyl) propionic acid-d4 sodium salt as an internal reference. The angle pulse and relaxation time were fixed at 90° and 30 s, respectively. The solid-state ¹³C-CP/MAS NMR spectra were obtained using an angle pulse of 37°, frequency of 12 kHz, contact time of 3 ms, and relaxation time of 3 s.

Morphology properties were studied in a scanning electron microscope (SEM, Shimadzu, model SS 550 Superscan). Energy spectra were analysed in a SEM-coupled energy dispersive spectroscope. Prior to SEM imaging, the hydrogels swollen to equilibrium were brought into liquid nitrogen in order to assure their original morphology. Then, the frozen samples were lyophilized for 72 h and subsequently sputter-coated with a thin layer of gold. SEM images were obtained by applying an acceleration voltage of 15 kV and a current intensity of 30 μA. In order to gain further insight into the morphology, the samples were imaged in a transmission electron microscope (TEM) in which it is possible to see small details. Bright field TEM images were obtained in a JEM-1400 microscope (JEOL) by applying an acceleration voltage of 120 kV. Before TEM imaging, an aliquot of a stirred suspension of samples in isopropyl alcohol was added dropwise onto a 200 mesh copper grid covered with a thin layer of carbon.

Results and discussions

Characterization of vinylpectin and vinyltitania

The chemical modification of pectin is an efficient strategy to convert uncross-linked polysaccharides into hydrogel. Covalent cross-links can be added by reacting vinyl monomers with the vinyl groups of functionalized polysaccharide. In order to prevent loss of titania from the hydrogel into the surrounding liquid, the nanoparticles were also vinyl-functionalized so that the binding formed with the polysaccharide is stable enough. The reactions were monitored by ¹H NMR and solid-state ¹³C-CP/MAS NMR spectroscopy.

Fig. 1a shows the ¹H NMR spectra of pectin, vinylpectin, and GMA (as the reference). The signals observed at 6.18 and 5.74 ppm in the spectra of vinylpectin are related to vinyl carbon-linked hydrogen derived from GMA. The signal at 1.95 ppm was attributed to the hydrogen of methyl groups in the vinyl carbons. The signals that appear at the spectral region of 3.55–4.40 ppm were associated to the glyceryl spacer. The results are strong evidence of the methacrylate conjugation of pectin with GMA.^28,29

Fig. 1 ¹H NMR spectra of pectin, vinylpectin and GMA (left) and ¹³C-CP/MAS NMR spectrum of vinyltitania (right).

The signals that appeared at 16.8 and 169.2 ppm in the ¹³C-CP/MAS NMR spectrum of vinyltitania (Fig. 1b) were assigned to the methyl carbon and carbonyl groups, respectively. However, as this technique is not sensitive enough to generate well-defined signals, but only broad signals, it is not possible to identify other carbons from the original spectrum. In order to have better insight into the functionalization of titania, the broad signals were deconvoluted from original signals to remove the widening effect and resolve the set of overlapping peaks. The signal at 140–120 ppm was decomposed into two peaks: 137.6 and 127.6 ppm, being both related to vinyl carbon groups. The signal at 90–63 ppm was resolved into three peaks: 67.5, 77.9 and 84.9 ppm, all of them associated with the glyceryl spacer. A downfield shift of ca. +10 ppm was observed for the C1, C2 and C3 carbons in the glyceryl spacer on titania in relation to their usual positions in organic compounds, which suggests that GMA bonded onto titania by means of oxygen atoms.^30,31 The ¹³C-CP/MAS NMR spectra evidenced the coordination of the methacrylate conjugation of GMA onto titania.

Pore morphology

Polymer hydrogel morphology factors such as porosity and small surface details give an important understanding on water penetration into and solute release from the hydrogel. Such factors are of great interest since they are directly related to diffusion. In this line of reasoning, microscopic morphology of hydrogels is conceptually interesting and needs to be understood to address further applications. SEM is an important tool that provides the advantage of direct visualization of the sample topography, mapping fracture or surface features at a microscopic level. Fig. 2 shows the micrographs of hydrogels with different amounts of vinylpectin, HEMA and vinyltitania. The hydrogels are shown to be wholly porous, independently of the amount of polymers or titania. Porous morphology can play an important role in water absorption by improving swelling performance. This characteristic allows a drug to diffuse into the swollen polymer network through pores, and this makes the hydrogels suitable for use in release systems and the targeted delivery of drugs.

Fig. 2 Micrographs of hydrogels with different amounts of pectin, HEMA and titania, taken from samples that were freeze-dried after being swollen to equilibrium in water.

In order to gain additional insight into actual morphology, the sample was swollen to equilibrium and subsequently fragmented prior to TEM imaging. The TEM images of the hydrogel nanocomposite 30-70-0.5 are shown in Fig. 3a and b. The images show spherical nanoparticles in the hydrogel. In addition to titania morphology, other relevant information that can be taken from the images is that the nanoparticles did not diffuse outwards from the porous hydrogel after being swollen. In such a case, possible interactions between the hydrogel and titania are indicated. The evidence that titania is, in fact, in the nanocomposite (Fig. 3c) was shown by EDS (Fig. 3d). The signal relative to Ti appeared at 4.7 eV.

Fig. 3 (a and b) TEM and (c) SEM images of the hydrogel nanocomposite 30-70-0.5. (d) EDS analysis of (c).

Swelling properties

The pure hydrogels and nanocomposites were swollen in buffer solutions at pH 2, 7 and 10 at 37 °C, and the experimental data are shown as equilibrium swelling degree (%) in Fig. 4. This approach allows the evaluation of the capacity of the hydrogels to absorb and retain water and its dependence on composition and pH. Pectin-only hydrogel proved to be extremely brittle when swollen, and thus the analysis of this sample was not taken further. The swelling degree clearly decreased with increasing HEMA. The addition of HEMA leads to a more densely cross-linked polymer network. As a result, the hydrogel does not expand enough to allow greater water absorption. This explains why the hydrogel (30-70), which contains a larger amount of HEMA, showed lower absorption capacity.

Fig. 4 Equilibrium swelling degree for pure hydrogels and nanocomposites at the indicated pH at 37 °C. The experimental data were obtained as the average over the two swelling determinations (n = 2). Columns with error bars represent mean ± standard deviation.

There was a decrease in the water absorption potential upon nanoparticle addition. Titania is believed to be a factor restricting, to some extent, the motion of polymer chains, so that water hardly diffuses into the hydrogel. The leading cause of that can be related to the interactions between the nanoparticles and the polymers. Gaharwar et al.¹⁹ reported that silica nanoparticles significantly influenced the swelling capacity of a PEG hydrogel as a result of polymer–nanoparticle interactions. In the present case, the tendency to reduce water absorption potential upon titania addition, which was observed in all compositions, is thought to be a consequence of highly effective interactions resulting from covalent bonds between the polymers and the nanoparticles.

PolyHEMA hydrogel has been demonstrated to be non-responsive to pH, owing to non-ionic characteristics of this polymer. However, systems combining polyHEMA with electrolyte-containing substances undergo pH-induced structural changes.³²

In the current study, the association of HEMA with pectin allowed the formation of a hydrogel with the ability to undertake swelling changes in response to pH. The dependence of swelling degree on pH was attributed to the –COOH groups (bearing electrolyte) of pectin. In solutions with pH higher than 3.55–4.10 (pK_a of pectin), the –COOH groups are in the –COO– form. Under this condition, anion–anion electrostatic repulsion forces are developed in the polymers, leading to hydrogel expansion, which makes water absorption easier. On the other hand, such forces are weakened at low pH, owing to protonated COOH groups, thus affecting the polymer expansion and consequently the swelling.

Mechanical properties

The successful application of hydrogels as controlled release systems, which are often used in long-term therapy, is related to their mechanical characteristics. In the case of failure or significantly reduced mechanical strength, the integrity of the hydrogel under given conditions is affected.

In order to gain important insights into their mechanical behavior, the samples were examined in terms of modulus of elasticity (E) and apparent cross-linking density (ν_e). E is obtained when the material undergoes an elastic, or nonpermanent, deformation. This means that when the gel is subjected to a compressive load, the deformation is accommodated by the reorganization of the polymer chains inside the gel. When the applied load is released, the chains return to their original configurations. The elastic deformation corresponds to the linear portion of the curve of stress vs. strain.³³

E was obtained from the linear slope (up to 20% of strain) of the stress–strain curves (Fig. 5a and b), plotted according to eqn (2). ν_e was also determined from the slope of the linear dependence of stress vs. strain, but by plotting according to eqn (3). The adjusting parameters are summarized in Table 2.

Fig. 5 Applied stress versus strain of hydrogels with (a) and without (b) titania.

Table 2 Values of elastic modulus, E, and apparent cross-linking density, ν_e, for hydrogels with and without titania

Hydrogel	E (×10^−3 MPa)	νe (×10^−5 mol cm^−3)
30-70	10.04 ± 0.12	4.37 ± 0.40
30-70-0.5	12.14 ± 0.86	4.88 ± 0.15
50-50	8.50 ± 0.31	3.21 ± 0.27
50-50-0.5	9.30 ± 0.24	4.06 ± 0.32
70-30	7.10 ± 0.04	3.50 ± 0.10
70-30-0.5	8.10 ± 0.10	3.88 ± 0.09

---

The pure hydrogels experienced an increase in E when the amount of HEMA was increased. HEMA showed a more important effect on mechanical behavior than pectin. For example, the hydrogel 70-30, containing a larger amount of pectin, showed E value lower than the hydrogel 50-50. The greater the E values, the more inflexible the hydrogel. The increase in E is associated with an increase in ν_e, which was more evident in the HEMA-richer hydrogel. This means that a loader is needed for the sample being compressed at 1 mm depth. From the physical-chemical point of view, the larger amount of HEMA produced more cross-linked points and yielded a hydrogel with a tighter structure within which the polymer chains are very close to each other. This type of architecture allows the formation of non-covalent interactions among the polymers that also contribute to increased ν_e. Given that an increase in both E and ν_e is related to a decrease in the water absorption, it can be thought that the mechanical behavior of the hydrogels reflects their swelling degree.

The hydrogels showed a marked increase in E upon addition of titania. The hydrogel nanocomposites appeared to be stiffer than the pure hydrogels, since they required an additional compressive force to be deformed at 1 mm. Furthermore, titania played a significant role in ν_e, which meant a substantial increase in the cross-linking density. In such a case, the increase in ν_e is related not only to non-covalent polymer–polymer interactions but also to chemical bonds of vinyltitania with the hydrogel.

Vitamin B₁₂ release study

Fig. 6 shows the time-dependent release of Vit-B₁₂ from the pure hydrogels and nanocomposites at pH 2, 7 and 10 at 37 °C. The effect burst was studied by determining the amount of drug released (%) at the first 30 min (t₃₀) and at the equilibrium stage (t_∞). In the most elementary form, t₃₀ corresponds to the release profile at early times, and t_∞ means time necessary to release to reach equilibrium. The values of each parameter were obtained from release curves of Fig. 6, and the data are shown in Table 3.

Fig. 6 Time-dependent Vit-B₁₂ release curves of pure hydrogels and nanocomposites at the indicated pH at 37 °C.

Table 3 Kinetics parameters of the release from pure hydrogels and nanocomposites at pH 2, 7 and 10: release percentages at the first 30 min (t₃₀) and at the equilibrium stage (t_∞)

Vit-B12 release (%)
Sample	pH 2		pH 7		pH 10
	t30	t∞	t30	t∞	t30	t∞
30-70	5.57	29.82	6.76	45.03	6.77	52.41
30-70-0.5	2.21	16.34	3.78	32.33	5.22	42.33
50-50	6.08	22.66	6.39	41.31	3.54	48.00
50-50-0.5	3.22	21.19	3.54	35.44	2.82	43.69
70-30	9.36	43.35	9.13	47.66	5.28	62.16
70-30-0.5	5.34	36.1	4.6	35.19	3.24	54.91

---

Values of t_∞ showed that Vit-B₁₂ release from the hydrogels changed in response to pH (Fig. 4). As a general trend, t_∞ increases with increasing pH, suggesting sustained release characteristics.

pH triggered Vit-B₁₂ release, but it did not influence the burst effect, which titania did. Values of t₃₀ demonstrated that the initial burst release from the hydrogel 30-70 was reduced by ca. 60% upon addition of titania. A low initial burst was also found for other nanocomposites, compared with the respective pure hydrogels.

Studies of release mechanisms and release kinetics by diffusive-based models

With the purpose of predicting whether diffusion is the leading factor in our systems, the more general version of the power law equation Korsmeyer–Peppas (eqn (4)) was used to adjust the experimental data, since it provides a significant understanding on the release mechanism. This semi-empiric equation is viewed to be the most comprehensive mathematical model used to determine the drug release profile of water-swellable polymers. It describes the drug release rate as directly proportional to the square root of time.^34–36where M_t and M_∞ are the mass of solute released from hydrogel at a specific time and at equilibrium, respectively; n is the diffusion coefficient; and k is the geometric and structural characteristic of the hydrogel.

The applicability of eqn (4) is limited to the first 60% of the released solute (M_t/M_∞ < 0.60), when the drug release is proportional to time. Weibull function is an empirical equation (eqn (5)) that has been used to describe the process and is valid for the entire release profile.³⁵

where d and k′ are constants. The k′ parameter provides important information on release kinetics, since it is closely related to the rate constant of release.

Analysis of kinetics parameters for Vit-B₁₂ release

The n parameter is related to the geometrical shape of the device. Although hydrogels can be prepared in different shapes, such as spheres, cylinders, and discs, we decided on a cylindrical form. In such a case, the following relation of n values to the release mechanism is considered: n = 0.45 means Fickian diffusion, which suggests that drug diffusion rate is slower than relaxation rate; 0.45 < n < 0.89 indicates an anomalous transport that is an intermediate state between Fickian diffusion and relaxation; n = 0.89 corresponds to a case II transport in which the release is regulated by macromolecular relaxation and is independent of time; and n > 0.89 for super case II transport, the contribution of diffusion, macromolecular relaxation, and erosion of the polymer. The values of n were obtained from slopes of the logarithmic curves of M_t/M_eq (<0.60) vs. time according to eqn (4), and the adjusting parameters are shown in Table 4.

Table 4 Adjusting parameters of eqn (4) and (5) to Vit-B₁₂ release from the hydrogels at pH 2, 7 and 10^a

Sample	pH 2		pH 7		pH 10
	n^b	k′^c (h)	n^b	k′^c (h)	n^b	k′^c (h)
a The fitting parameters were estimated by a least-squares approach, with greater than or equal to a 95% level of confidence (Origin 8.0).b Parameter obtained from the Peppas equation (eqn (4)).c Parameters obtained from the Weibull equation (eqn (5)).
30-70	0.65 (±0.03)	0.219 (±0.006)	0.65 (±0.02)	0.194 (±0.019)	0.59 (±0.01)	0.117 (±0.011)
30-70-0.5	0.63 (±0.03)	0.145 (±0.015)	0.62 (±0.01)	0.116 (±0.011)	0.57 (±0.01)	0.054 (±0.013)
50-50	0.73 (±0.02)	0.364 (±0.019)	0.78 (±0.02)	0.125 (±0.006)	0.79 (±0.03)	0.059 (±0.004)
50-50-0.5	0.61 (±0.03)	0.312 (±0.021)	0.63 (±0.03)	0.121 (±0.014)	0.75 (±0.03)	0.026 (±0.007)
70-30	0.60 (±0.02)	0.178 (±0.021)	0.72 (±0.03)	0.237 (±0.041)	0.78 (±0.04)	0.042 (±0.021)
70-30-0.5	0.56 (±0.02)	0.099 (±0.008)	0.64 (±0.01)	0.106 (±0.014)	0.66 (±0.03)	0.027 (±0.004)

---

The release mechanism of Vit-B₁₂ from the pure hydrogels was found to be anomalous. This means that a combination of Fickian diffusion and macromolecular relaxation appears to be the main driving force for release. The same release mechanism was observed for the nanocomposites, but with a tendency towards Fickian diffusion (n → 0.45). Titania causes an attenuation in the polymer motion and consequently in the macromolecular relaxation. Pore tortuosity also appears to be another factor affecting the anomalous release, as a result of distribution of the nanoparticles during the preparation of the nanocomposite.

k′ was obtained by fitting the experimental data of release to eqn (5), and the adjusting parameters are shown in Table 4. The values of k decreased upon addition of titania. This decrease is associated with the reduction in the initial release burst, because k′ is related to the release rate constant.

In vitro cell culture toxicity assays

The successful application of the hydrogels in the biomedical field is dependent on their compatibility with cells (pharmacological potential). The safety of the proposed materials that are to be used as a controlled drug release system was studied by determining cytotoxicity. The cytotoxic concentrations for 50% VERO cells (CC₅₀) were calculated as the dose necessary to inhibit the cell growth by 50%, for a cut-off of 1000 μg mL⁻¹, as shown in Table 5. The data were also shown in terms of % growth inhibition (GI). The values of CC₅₀ were found to be higher than 1000 μg mL⁻¹, and the values of GI were much lower than 50%. This means that the samples did not inhibit cell growth, demonstrating biocompatibility.

Table 5 In vitro cell culture toxicity of hydrogel nanocomposites

Samples	CC50 (μg mL^−1)	% growth inhibition ± SD at 1000 mg mL^−1
30-70	>1000	1.90 ± 0.25
30-70-0.5	>1000	0.00 ± 0.00
50-50	>1000	2.90 ± 0.13
50-50-0.5	>1000	8.77 ± 1.25
70-30	>1000	2.93 ± 0.89
70-30-0.5	>1000	2.67 ± 0.35

---

Conclusions

Hydrogel nanocomposites of pectin, HEMA and titania for Vit-B₁₂ controlled release with reduced initial release burst were prepared. Vinyl groups were added into pectin as a covalent approach to convert uncross-linked polysaccharides into hydrogel. To prevent loss of titania from the porous hydrogel into water, the nanoparticles were also vinyl-functionalized for further reaction with vinylpectin. Titania provided hydrogels with improved mechanical properties and reduced initial burst release, which is related to a decrease in the release rate constant. The release mechanism of Vit-B₁₂ from the pure hydrogels was driven by Fickian diffusion and macromolecular relaxation. Upon addition of titania, the contribution of the macromolecular relaxation was minimized, suggesting a tendency towards Fickian diffusion. Cytotoxicity tests showed that both the pure hydrogels and the nanocomposites are biocompatible. The results of the current work are relevant for the development of advanced nanobiomaterials for use in biomedical applications as a controlled drug release system.

Acknowledgements

E. P. S. thanks the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for the doctorate fellowship. M. R. G. thanks the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for the post-doctorate fellowship (PDS). A. F. R. and M. H. K. acknowledge the financial supports given by CNPq, CAPES.

References

1. S. W. Shalaby and K. J. L. Burg, Absorbable and Biodegradable Polymers: Cyanoacrylate-Based Systems as Tissue Adhesives, CRC Press, New York, 2004.
2. E. S. Dragan, Chem. Eng. J., 2014, 243, 572–590.
3. D. Das and S. Pal, RSC Adv., 2015, 5, 25014–25050.
4. G. R. Mahdavinia, H. Etemadi and F. Soleymani, Carbohydr. Polym., 2015, 128, 112–121.
5. E. Bakaic, N. M. B. Smeets and T. Hoare, RSC Adv., 2015, 5, 35469–35486.
6. J. Liu, L. Cui, N. Kong, C. J. Barrow and W. Yang, Eur. Polym. J., 2014, 50, 9–17.
7. M. Boruah, P. Gogoi, A. K. Manhar, M. Khannam, M. Mandal and S. K. Dolui, RSC Adv., 2014, 4, 43865–43873.
8. E. T. Tenório-Neto, M. R. Guilherme, M. K. Lima-Tenório, D. B. Scariot, C. V. Nakamura, A. F. Rubira and M. H. Kunita, Colloid Polym. Sci., 2015, 293, 3611–3622.
9. L. Ye, Y. Tang and D. Qiu, Colloids Surf., A, 2014, 447, 103–110.
10. A. S. Hasan, M. Socha, A. Lamprecht, F. E. Ghazouani, A. Sapin, M. Hoffman, P. Maincent and N. Ubrich, Int. J. Pharm., 2007, 344, 53–61.
11. M. Swierczewska, H. S. Han, K. Kim, J. H. Park and S. Lee, Adv. Drug Delivery Rev., 2015 DOI:10.1016/j.addr.2015.11.015.
12. W. E. Hennink and C. F. van Nostrum, Adv. Drug Delivery Rev., 2012, 64, 223–236.
13. H. Hezaveh and I. Muhamad, J. Mater. Sci.: Mater. Med., 2013, 24, 1443–1453.
14. X. Huang and C. S. Brazel, J. Controlled Release, 2001, 73, 121–136.
15. H. Takabe, M. Ohkuma, Y. Iwao, S. Noguchi and S. Itai, Pharmacol. Pharm., 2013, 4, 578–583.
16. K. Fu, R. Harrell, K. Zinski, C. Um, A. Jaklenec, J. Frazier, N. Lotan, P. Burke, A. M. Klibanov and R. Langer, J. Pharm. Sci., 2003, 92, 1582–1591.
17. A. R. Ahmed, K. Elkharraz, M. Irfan and R. Bodmeier, Pharm. Dev. Technol., 2010, 17, 66–72.
18. R. A. Hule and D. J. Pochan, MRS Bull., 2007, 32, 354–358.
19. A. K. Gaharwar, C. Rivera, C.-J. Wu, B. K. Chan and G. Schmidt, Mater. Sci. Eng., C, 2013, 33, 1800–1807.
20. F. Sun, H. Zhou and J. Lee, Acta Biomater., 2011, 7, 3813–3828.
21. J. Yang, C.-R. Han, J.-F. Duan, F. Xu and R.-C. Sun, J. Phys. Chem. C, 2013, 117, 8223–8230.
22. D. Pasqui, A. Atrei, G. Giani, M. De Cagna and R. Barbucci, Mater. Lett., 2011, 65, 392–395.
23. A. Kubacka, C. Serrano, M. Ferrer, H. Lünsdorf, P. Bielecki, M. L. Cerrada, M. Fernández-García and M. Fernández-García, Nano Lett., 2007, 7, 2529–2534.
24. E. P. Silva, M. R. Guilherme, E. T. Tenório-Neto, A. F. Rubira, M. H. Kunita, L. Cardozo-Filho and C. G. Alonso, Mater. Lett., 2014, 136, 133–137.
25. K. S. Suslick and G. J. Price, Annu. Rev. Mater. Sci., 1999, 29, 295–326.
26. M. R. Guilherme, G. M. Campese, E. Radovanovic, A. F. Rubira, J. P. A. Feitosa and E. C. Muniz, Polymer, 2005, 46, 7867–7873.
27. P. Skehan, R. Storeng, D. Scudiero, A. Monks, J. McMahon, D. Vistica, J. T. Warren, H. Bokesch, S. Kenney and M. R. Boyd, J. Natl. Cancer Inst., 1990, 82, 1107–1112.
28. A. P. Gerola, D. C. Silva, S. Jesus, R. A. Carvalho, A. F. Rubira, E. C. Muniz, O. Borges and A. J. M. Valente, RSC Adv., 2015, 5, 94519–94533.
29. W. Wang, L. Zong and A. Wang, Int. J. Biol. Macromol., 2013, 62, 225–231.
30. A. Foris, Magn. Reson. Chem., 2000, 38, 1044–1046.
31. E. M. Rockafellow, X. Fang, B. G. Trewyn, K. Schmidt-Rohr and W. S. Jenks, Chem. Mater., 2009, 21, 1187–1197.
32. S. L. Tomić, M. M. Mićić, S. N. Dobić, J. M. Filipović and E. H. Suljovrujić, Radiat. Phys. Chem., 2010, 79, 643–649.
33. W. D. Callister, Materials Science and Engineering: An Introduction, John Wiley, New York, 1940, p. 138.
34. J. Siepmann and N. A. Peppas, Adv. Drug Delivery Rev., 2001, 48, 139–157.
35. D. Costa, A. J. Valente, M. G. Miguel and J. Queiroz, Langmuir, 2011, 27, 13780–13789.
36. M. A. Lauzon, E. Bergeron, B. Marcos and N. Faucheux, J. Controlled Release, 2012, 162, 502–520.

---