Novel hybrid formulations based on chitosan and a

0 downloads 0 Views 2MB Size Report
Mar 23, 2017 - ity in organic solvents, low mechanical strength and poor chemical resistance in acidic ... In some situations the dissolution or degradation of CH is ... Materials. Chitosan (CH) of medium molecular weight, ciprofloxacin (CPF),.
Reactive and Functional Polymers 114 (2017) 118–126

Contents lists available at ScienceDirect

Reactive and Functional Polymers journal homepage: www.elsevier.com/locate/react

Novel hybrid formulations based on chitosan and a siloxane compound intended for biomedical applications Irina Elena Bordianu-Antochi ⁎, Mihaela Olaru, Corneliu Cotofana “Petru Poni” Institute of Macromolecular Chemistry, 41A Grigore Ghica Voda Alley, 700487 Iasi, Romania

a r t i c l e

i n f o

Article history: Received 3 December 2016 Received in revised form 28 February 2017 Accepted 22 March 2017 Available online 23 March 2017 Keywords: Chitosan Hybrid materials Ciprofloxacin Drug delivery Schott model

a b s t r a c t The present study reports on the obtaining of chitosan/siloxane–based microspheres by coacervation/precipitation method and their use as efficient drug delivery vehicles for ciprofloxacin, one antibacterial synthetic drug belonging to fluoroquinolones group. These new hybrid formulations were analyzed in terms of structural characterization (FTIR, SEM, TG, DSC techniques), swelling capacity and in vitro drug release. The release mechanism of the model drug was investigated by means of several kinetic models, i.e., zero order, first order, Higuchi model, Korsmeyer–Peppas model, Hixson–Crowell model, Baker–Lonsdal model, Weibull model and Schott model. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Despite its intensive use in different fields of research and application, chitosan (CH) continues to awaken the researcher's interest due to its outstanding physicochemical and biological characteristics. Being one of the most abundant natural polysaccharides, CH is a positively charged natural polymer, is biocompatible, biodegradable, non–toxic and presents antibacterial properties [1]. Due to its coagulation ability and immuno–stimulating activity [1], CH is widely used with success in various domains, such as tissue engineering, wound healing, food additives, drug delivery, cancer diagnosis, water treatment or cosmetics [2]. The presence of functional groups (like hydroxyl, amino or N–acetyl) allows CH to self-assemble into a variety of supramolecular structures either through a chemical reaction [3] or via the so–called ISISA process (Ice Segregation Induced Self Assembly process), which is based on the unidirectional freezing of a hydrogel material [4]. The self–assemble ability offers to CH carriers a tremendous potential with application in fields, such as peptide/protein delivery, vaccine delivery, controlled drug release [2], gene delivery [5,6]. Beside its special properties, CH presents also some drawbacks connected with the poor solubility in organic solvents, low mechanical strength and poor chemical resistance in acidic media [7]. In order to avoid these types of issues CH structure is widely modified with different compounds, such as natural polymers (e.g. alginate, starch, cellulose), synthetic polymers (e.g. ⁎ Corresponding author at: Institute of Macromolecular Chemistry “Petru Poni” Iasi, 41A Grigore Ghica Voda Alley, 700487 Iasi, Romania. E-mail addresses: [email protected] (I.E. Bordianu-Antochi), [email protected] (M. Olaru), [email protected] (C. Cotofana).

http://dx.doi.org/10.1016/j.reactfunctpolym.2017.03.013 1381-5148/© 2017 Elsevier B.V. All rights reserved.

poly(ethylene glycol), poly(ethylene oxide), poly(Ɛ-caprolactone), poly(ethylene glycol)–co–poly(lactone)diacrylate) [2] or silica–based compounds [8–10]. In some situations the dissolution or degradation of CH is avoided also by the addition of cross–linking agents such as glutaraldehyde, formaldehyde, [10] epichlorhydrine [11], dialdehyde starch, sulphuric acid [9], etc. Taking into consideration the CH properties and the advantages that a silica–based compound could offer (large surface area, enhanced mechanical resistance, improved physical properties), a special attention was given to the obtaining of CH/siloxane hybrid materials due to improved properties and enhanced application fields [10,12–14]. The aim of the present paper was to obtain new formulations based on CH/siloxane hybrid materials intended for the delivery of a model drug, i.e., ciprofloxacin (CPF) which is a second generation quionolone efficient against gram positive and negative bacteria. This drug is widely used to treat conjunctivitis, keratitis, gonorrhoea, osteomyelitis, infections of the urinary and respiratory tracts, low respiratory tract infections, a.s.o [15]. The hybrid materials described in this work were obtained in the form of hydrogel microspheres using the coacervation/ preparation method and were further dried by freeze–drying technique. The mechanism release of the model drug was investigated by means of several mathematical models, i.e., zero order, first order, Higuchi model, Korsmeyer–Peppas model, Hixson–Crowell model, Baker–Lonsdal model, Weibull model and Schott model. Although several papers presented the preparation of drug delivery formulations based on CH microspheres [11,16–20] to our knowledge this paper is the first one reporting the obtaining of chitosan microspheres through coacervation/precipitation method incubated with a model drug (an antibiotic) and coated with a siloxane compound. Most often, the drug delivery

I.E. Bordianu-Antochi et al. / Reactive and Functional Polymers 114 (2017) 118–126

formulations based on CH modified with siloxane compounds involved the use of CH in the form of films or membranes [21–23]. 2. Experimental section 2.1. Materials Chitosan (CH) of medium molecular weight, ciprofloxacin (CPF), 98% (HPLC), 3–(chloropropyl)–trimethoxysilane (CPTMS) with a purity of 97%, 1–ethyl–3–(3–dimethylaminopropyl) carbodiimide hydrochloride (EDC), N–hydroxysuccinimide (NHS) and sodium hydroxide pellets were purchased from Sigma Aldrich. 2.2. Preparation method The preparation procedure of the hybrid samples is schematically represented in Scheme 1. The CH microspheres were obtained using the coacervation/precipitation method [24]. Briefly, a solution of 2% CH was obtained by dissolving the CH powder in an aqueous solution of 1% acetic acid and the solution was stirred for 24 h at room temperature. The CH microspheres were obtained by dropping the CH solution in sodium hydroxide (NaOH, 3 M) using a needle and after precipitation the microspheres were left under very slow agitation for another 1 h. In order to remove any alkali traces the CH microspheres were washed with distilled water until the pH = 7. In the second step, a CPF aqueous solution (1%) was prepared and left under vigorous stirring for several hours until solubilisation was achieved. It is well known that CPF is one of the weakly basic drugs and is insoluble in water at neutral pH. For this reason sulfuric acid was used (at pH = 5.5–6) in order to

119

facilitate the solubility of the drug. The coupling between CH and CPF was favoured by coupling agents, 1–ethyl–3–(3–dimethylaminopropyl) carbodiimide hydrochloride/N–hydroxysuccinimide (EDC/NHS). The molar ratio CPF:EDC:NHS was 1:1:1 and the CH microspheres, in hydrogel form, were transferred in this system and left under stirring at 40 °C for 24 h. In the third step an amount of CPTMS (1%) was placed in an acidic aqueous solution and left under stirring at 40 °C for 6 h. The CH microspheres containing CPF were washed with distilled water, transferred in this solution and left under stirring for other 24 h. In the final step, the hybrid materials containing CPF were washed with distilled water and dried by freeze–drying technique. Both unmodified and CH microspheres containing CPF were freeze–dried and subjected to structural characterization, too. 2.3. Drug entrapment efficiency The drug entrapment was calculated using 50 mg of microspheres crushed in a glass mortar and the powder obtained was introduced in 100 ml of HCl solution 0.1 N and kept under stirring for 4 h. In the next step, the solution was filtered and the drug content was analyzed following the evolution of the band located at 277 nm using a UV–VIS spectrophotometer. The entrapment efficiency was determined using formula (1) [25]. The quantity of drug entrapped in the microspheres resulted to be 40% of the initial quantity.

Drug Entrapment % ¼

Actual drug content  100 Theoretical drug content

Scheme 1. Schematic representation of the hybrid microspheres preparation.

ð1Þ

120

I.E. Bordianu-Antochi et al. / Reactive and Functional Polymers 114 (2017) 118–126

3. Materials characterization The hydrogel microspheres were dried by freeze-drying technique using a Martin Christ ALPHA 1–2LD Freeze–drier for 3 days. The infrared spectra were recorder using a Bruker Vertex 70 device at room temperature. Spectra were recorded in the 400–4000 cm−1 region with 64 scans, using the KBr pellet technique and the Opus 5 FTIR Software. The morphology of the samples was investigated using a Quanta 200 scanning electron microscope and the thermal stability of the samples was studied with a STA 449 F1 Jupiter device (for TGA measurements) and a differential scanning calorimeter SC 200 F3 Maia (for DSC measurements). In both cases samples were placed in aluminium crucibles and the experiments were recorder in nitrogen atmosphere. The UV– VIS Specord M42 spectrophotometer was utilized for the investigation of the drug release behaviour. 4. Results and discussion 4.1. Fourier transform infrared spectroscopy (FTIR) Fig. 1 displays the FTIR spectra for neat CPF, neat CH microspheres, CH microspheres containing CPF and the spectrum for hybrid materials. Analyzing the corresponding spectra of neat CPF and CH microspheres, the characteristic peaks were easily identified. Thus, in the neat CPF spectrum the broad band between 3550 and 3350 cm−1 corresponds to intermolecular hydrogen bonding and to OH group. The absorption bands located between 3000 and 2845 cm− 1 are assigned to the stretching vibration of C\\H bonds from alkyl and aromatic groups, while the stretching vibration of the C_O group is confirmed by the peak from 1730 cm−1 [26]. The prominent peak from 1617 cm−1 appears due to N\\H bending vibration of quinolones and those between 1500 and 1450 cm−1 are attributed to O\\C\\O bonds from carboxylic group. The bending vibration of OH group from the same carboxylic group is confirmed by the presence of the absorption bands between 1310 and 1260 cm−1, while the peak positioned at 1036 cm−1 proves the presence of C\\F group [27,28]. FTIR spectrum of neat CH microspheres shows the presence of a broad band situated between 3500 and 3300 cm−1 that can be attributed to both OH and NH2 groups stretching vibrations, since these bands overlap one another. The absorption bands observed at 3000 cm− 1 and 2893 cm−1 can be ascribed to the stretching vibrations of \\C\\H bonds from alkyl groups [29]. Bands from 1654 cm−1 can be associated to amide I region, which involves the contribution of C_O stretching

group, while the ones from 1561 cm− 1 can be related to amide II bands, i.e., the stretching of\\NH group from free amine groups. Moreover, the peaks located between 1438 and 1350 cm−1 illustrate the presence of the bending vibrations of \\C\\H and \\CH2 groups [30], while the ones from 1240 cm− 1 to 911 cm− 1 are characteristic to stretching vibrations of C\\O or C\\O\\C bonds [31]. Although most of the absorption bands of CH microspheres containing CPF and of the hybrid materials are overlapping, several particularities related to the shift of the absorption bands toward higher wavenumbers, as well as the absence or presence of some significant bands can be evidenced. As follows, in the FTIR spectrum of CH containing CPF, the shift of amide II band to higher wavenumbers (from 1561 to 1578 cm−1), the absence of C_O absorption band from CPF (initially identified at 1730 cm− 1) and the presence of the peak recorded at 1030 cm−1 corresponding to C\\F stretching vibration show that the drug was incubated in the CH microspheres structure [32,33]. Moreover, in the FTIR spectrum corresponding to hybrid sample the absorption band between 1100 and 1000 cm−1 can be assigned to C\\O\\Si bonds [34], while the bands positioned at 1156 cm−1 [9] and between 500 and 450 cm−1 can be attributed to Si\\O\\Si bonds [35]. The presence of these absorption bands proves the interactions between chitosan and siloxane compound. 4.2. Scanning electron microscopy (SEM) The morphology of the samples was investigated by SEM technique. In Fig. 2 are displayed images of a whole microsphere, external and internal surface corresponding to neat CH microsphere (a), CH microsphere with CPF (b) and to the hybrid materials (c), respectively. In every step of the preparation method the samples present a spherical shape. The neat CH microspheres, Fig. 2a, present an irregular porous structure, while in Fig. 2b and c different morphologies can be distinguished. Thus, both CH microspheres containing CPF and the hybrid microspheres present at the external surface an ordered macroporous structure formed from nearly hexagonal pores with different sizes, ranging from 5 to 20 μm. The appearance of the hexagonal pores occurs due to a self–assembly process and is more prominent at the external surface and less at the internal one. A deeper analysis of Fig. 2 evidences that the self–assembly process is rather influenced by chemical or physical processes that took place between CH and CPF [4] and not by the freeze–drying process used to dry the samples [3]. In the case of the hybrid materials, siloxane compound forms clusters on the surface of the microspheres which almost fill the hexagonal pores and, most probably, is involved in the matrix protection and delay of the drug delivery. 4.3. Thermogravimetric analysis (TG/DTG) and differential scanning calorimetry (DSC) TG/DTG (Fig. 3) and DSC (Fig. 4) techniques were used to characterize the thermal behaviour of the samples and in the same time to obtain more information concerning CPF distribution in the hybrid microspheres.

Fig. 1. FTIR spectra for: neat CPF, CH microspheres, CH microspheres containing CPF and hybrid samples.

4.3.1. TG/DTG data TG and DTG curves (Fig. 3a and b) were recorded for neat CPF and for each step of the preparation method. The curves registered for neat CPF show a decomposition process based on three degradation stages, the data obtained being in accordance with those reported in literature [26]. The mass loss and the decomposition process starts at around 280 °C and all three degradation stages correspond to exothermic processes. At the end of the process, the residue was found to be around 20%. For the neat CH microspheres, there is a small weight loss below 100 °C that can be related to the loss of adsorbed water and to the breakdown of the hydrogen bonds formed between water and CH. The main thermal degradation process corresponds to the decomposition of CH chains and occurs between 240 and 593 °C, with a Tpeak at 293 °C. At

I.E. Bordianu-Antochi et al. / Reactive and Functional Polymers 114 (2017) 118–126

121

Fig. 2. SEM images of whole sphere, external and internal surface for: (a) CH microspheres; (b) CH microspheres with CPF; (c) hybrid samples.

the end of the process, which was recorded at around 600 °C, the quantity of the residue was 23.27%. The thermal stability of CH microspheres containing CPF is higher than those of neat CPF and of CH ones, the quantity of residue being of 31%. In this latter case, beside the weight loss below 100 °C, two exothermic peaks were recorded, i.e., one at Tpeak 251 °C, which corresponds to CH decomposition and another one at 286.5 °C that can be related to CPF decomposition [36].

Analysis of TG/DTG curves corresponding to hybrid materials evidences the presence of four stages of thermal degradation. Besides the characteristic stage of CH degradation (the exothermic peak with Tpeak at around 253 °C), the occurrence of other three stages distinctive for the decomposition of CPF and the organic part of the siloxane compound can be observed. Thus, three exothermic peaks can be identified at 331.6 °C, 424.5 and 525.7 °C, respectively. The attendance of these peaks and the quantity of residue at the end of the process (48.21%)

Fig. 3. TG (a) and DTG (b) curves corresponding to CPF, CH microspheres, CH microspheres with CPF and hybrid samples.

122

I.E. Bordianu-Antochi et al. / Reactive and Functional Polymers 114 (2017) 118–126

identified in the DSC curve of the neat CH microspheres is at 306 °C and can be related to the thermal and oxidative decomposition of the natural polymer. As regards the DSC curve for the hybrid materials an endothermic peak appears at 82 °C, at a lower value as compared to the neat CH microspheres. This peak also corresponds to dehydration of the natural polymer and, at the same time, suggests that the percent of hydrogen bonds and the quantity of water are smaller in the hybrid materials [37]. The exothermic peak is recorded at 262 °C, also at a lower temperature due to the formation of new bonds between CH and the other two components, process that leads to the lowering of the percentage of hydrogen bonds and decreasing of the CH crystallinity and stability. The appearance of one high temperature endothermic peak at 337 °C connected to the decomposition of the organic and inorganic parts of the matrix (also including CPF) sustains the fact that presence of siloxane compound increases the thermal stability of the hybrid material. 4.4. Swelling behaviour of the samples Fig. 4. DSC thermograms corresponding to CH microspheres and to hybrid samples.

confirms the presence of CPF and siloxane compound in the hybrid microspheres. 4.3.2. DSC data The DSC data come to sustain the results obtained from thermogravimetric analysis. The DSC curve recorded for the neat CH microspheres presents two peaks characteristic to the natural polymer. The board endothermic peak that can be identified at 88 °C corresponds to the loss of water and CH dehydration. The appearance of the endothermic peak at a temperature below 100 °C proves that the hydrogen bounds are formed mainly between amino groups and water, the energy needed to break this type of bonds being low. It is known that CH is a natural polymer with a strong affinity to water. The hydrogen bonds that CH can form with water are intra–molecular and they form due to the presence of hydroxyl and amino group in CH structure [37]. The second peak

One of the most important properties of a hydrogel is the capacity to absorb large amounts of water or biological fluids. The swelling equilibrium is reached when the elastic forces are equal to the osmotic ones (the last ones being related to the affinity of the material vs. solvent), this balance encouraging the penetrant to entrance into the hydrophilic part. The hydrophilic character of a hydrogel has a major impact on its swelling behaviour. The presence of a high number of amine, carboxyl or hydroxyl groups is usually connected with the absorption of a higher amount of water or biological fluids that confers to a material a high hydrophilicity. Also, the swelling behaviour is closely related to the drug release and loading properties [11]. In the present paper, the swelling behaviour of materials was studied in different aqueous media, i.e., distilled water, HCl solutions (simulated gastric fluid – SGF, pH = 1.2) and phosphate buffer saline solution, (PBS, pH = 7.4). Freeze–dried microspheres (corresponding to the three stages of the preparation method) were immersed in the aqueous media at 37 °C and held until the swelling equilibrium was attained. At determined periods of time (30 min,

Fig. 5. Swelling behaviour: (a) neat CH microspheres, (b) CH microspheres with CPF, (c) hybrid samples.

I.E. Bordianu-Antochi et al. / Reactive and Functional Polymers 114 (2017) 118–126

123

Fig. 6. Neat CH microspheres in HCl solution (a); neat CH microspheres dissolved in HCl solution after a few minutes (b).

1 h, 2 h, 4 h, 6 h, 24 h, 48 h, 72 h) the swollen microspheres were removed from the aqueous medium, were carefully blotted with filter paper and weighted. In order to determine the swelling ratio, formula (2) was used, where Sw is the swelling ratio, Ws is the weight of the microspheres in swollen state and Wd is the weight of the microspheres in dry state [38]. Sw ¼

Ws −Wd Wd

ð2Þ

Analyzing the spectra corresponding to the swelling behaviour it can be observed that the swelling capacity of the samples increases with time passing (Fig. 5). The swelling behaviour is similar for all three aqueous media, which implies that swelling ratio increases until a certain point when the equilibrium state is reached. For the neat CH microspheres, the swelling behaviour (Fig. 5a) is represented only in distilled water and PBS solution due to the fact that in SGF the CH microspheres dissolved in b 30 min (Fig. 6). In the latter case, this behaviour can be correlated with the presence of amino/imine groups that become NH+ 3 upon protonation in acidic medium, thus facilitating the dissolution of CH microspheres. In the other two aqueous media the neat CH microspheres indicate a quite similar behaviour, with a slight increase in swelling capacity in distilled water as compared to PBS solution, since the protonation ratio of CH microspheres at high pH values is low. For the CH microspheres containing CPF (Fig. 5b), the swelling capacity in the three aqueous media increases and one can observe that the samples are sensitive to aqueous environments depending on the solution pH. Moreover, in SGF medium the samples do not dissolve anymore, on the contrary, in this medium the samples present the highest swelling capacity. The explanation may arise from the fact that at acidic

pH values (i.e., pH b 3), the swelling capacity of samples increases mainly due to the presence of unreacted amino and imino groups from CH and\\NH groups from CPF. Upon protonation, an electrostatic repulsion between the polymer chains is occurring and a high amount of water is allowed to penetrate into the hydrogel network. At the same time, the relaxation of the matrix chains takes place and the water uptake increases. Instead, at basic pH values (i.e., pH N 6) this kind of forces or processes disappear and, in consequence, the swelling capacity is lower [11,39]. As regards the swelling behaviour of the hybrid microspheres, from Fig. 5c results that it is quite similar to CH microspheres containing CPF; the samples are sensitive to the pH value, but the presence of the siloxane compound in their structure seems to determine a decrease of the swelling capacity. This behaviour will have, in the end, a beneficial effect in terms of drug release and will be presented in detail in next paragraphs. 4.5. In vitro drug release The in vitro release behaviour of CPF from CH microspheres containing CPF and from hybrid samples was studied in two media, i.e., SGF (pH = 1.2) and PBS buffer (pH = 7.4) solutions. An amount of 20 mg of freeze–dried samples were immersed in 40 ml of dissolution medium and, subsequently, the bottles were introduced in a mechanical shaking bath at 37 °C (50 rpm). At different time intervals, i.e., 30 min, 1 h, 2 h, 4 h, 6 h, 24 h, 48 h and respectively 72 h, 10 ml of solution were removed, filtered, followed by the analysis of the evolution of the band located at 277 nm using a UV–VIS spectrophotometer. The quantity of solution withdrawn was replaced every time with the same fresh quantity of dissolution medium. The drug release percent was calculated using formula (3) [40] and in Fig. 7 are displayed the drug release

Fig. 7. Drug release profiles of (a) CH microspheres with CPF and (b) of hybrid samples.

124

I.E. Bordianu-Antochi et al. / Reactive and Functional Polymers 114 (2017) 118–126

Fig. 8. Photos of the samples after a few hours of in vitro drug release tests in SGF (a) and PBS buffer solution (b).

profiles of the CH microspheres containing CPF (a) and of the hybrid samples (b) in the two aqueous media. Drug release % ¼

Amount of drug release at t  100 Initial loaded amount of drug

ð3Þ

As it can be noticed, the CPF release from CH microspheres containing CPF took place with a fast burst release in the first 30 min in both aqueous media, in SGF solution being higher (68.81%) than in PBS buffer solution (49.54%). In the next hours, the release became higher and, up to 72 h, 88% of the drug was released in PBS buffer solution and 93% in SGF. Analyzing Fig. 7b, one can observe that hybrid samples presented almost the same behaviour, the difference being related to the quantity of drug release in the same period of time. The hybrid samples present a lower burst release in both media (56.81% in SGF solution and 40% in PBS buffer solution) and the quantity of released drug is lower after 72 h (86.36% in SGF solution and 68.81% in PBS buffer solution). These results prove that the presence of the siloxane compound in hybrid samples is beneficial since it fills the CH hexagonal pores, thus slowing down the burst release and lowering the amount of drug released. The reason of the high burst release in both situations may be related to several factors. Among of the most important factors is the morphology of samples [41]. The macroporous structure, with pore sizes ranging from 5 to 20 μm formed after a self–assembly process, seems to allow the drug release from the surface and from inside the samples. Also, the high swelling capacity influences the release of the drug in the early stages, a higher swelling degree/capacity determining the release of a higher amount of drug [42]. In the present study, the swelling behaviour seems to be in good agreement with the drug release behaviour, i.e., the burst release is higher in SGF than in PBS solution in the same period of time. Another aspect that could be taken into consideration as concerns the burst release is the formation of some cracks on the hybrid sample's surface that could have a role in this regard [41]. Although the appearance of visible cracks on the surface of hybrid microspheres was registered only after several hours (Fig. 8), we tend to believe that they start to form from the first moments of immersion, influencing in this way the burst release process. Usually, the burst release process is regarded as a disadvantage due the negative effects that could appear in the in vivo conditions. On the other hand, there are also situations when the burst release can be desired, for example in pulsatile release, targeted delivery or wound healing. As compared to other results and methods available in literature [16,19,43,44], the burst release recorded in the case of the studied samples is much higher, but even so the preparation method reported in the present paper looks like an efficient one. The presence of the siloxane compound in the hybrid samples helps lowering down the drug release and makes the formulations suitable for potential prolonged drug release systems. Moreover, the preparation method is a very simple one, permits the control of porosity and diameter of the samples, and offers the possibility to obtain stable samples in mild conditions without using organic solvent or surfactants [45,46], the

only solvent used being water. Thus, even if this preparation method has some disadvantages, among which a higher burst release, it has many other advantages that make it attractive as compared to other preparation techniques.

4.6. Kinetics of drug release The drug release mechanism from hybrid samples was evaluated by fitting the experimental data into some mathematical models, as summarised in Table 1 [47,48]. These mathematical models are the most common and the most applied ones in order to determine the release mechanism of a drug from a delivery system. The selection of the model that best corresponds to the samples depends on the regression coefficient (R2) and on the diffusion exponent (n) values [47]. Depending on n value, the mechanism release types are classified into Fickian mechanism – Case I (diffusion controlled – the diffusion of the penetrant is slower than the polymer chains relaxations, n = 0.5), Fickian mechanism – Case II (relaxation controlled – the diffusion of the penetrant is faster than the polymer chains relaxation, n = 1) and non – Fickian mechanism or anomalous mechanism (the diffusion of the penetrant is comparable to the polymeric chains relaxation, 0.5 b n b 1). There are some cases when n value is lower than 0.5, situation in which the process is called Less Fickian mechanism. This mechanism is still a Fickian one, but the diffusion of the penetrant is slower than Table 1 Mathematical models applied in order to investigate the release mechanism of CPF. Mathematical model

Equation of the Graphical representation mathematical model

Zero order

Qt = Q0 + K 0 t

First order

log Q = log Q0 − Kt/2.303 Q = KHt1/2

Higuchi model

Korsmeyer–Peppas Mt/M∞ = Ktn

Percent of cumulative amount of drug release versus time. Percent of drug remaining versus time. Percent of cumulative drug amount of drug release versus square root of time. Log of percent of cumulative drug release versus log of time. Cube root of drug remaining versus time. Model formula versus time.

Hixson–Crowell

1/3 = KHCt Q1/3 0 − Qt

Baker–Lonsdal

3/2[1 − (1 − F)2/3]-F = Kt log [−ln(1 − m)] = Log of percent of cumulative drug b log(t − Tt) − log a release versus log of time. Linear plot between t/Mt. and time. dMt/dM∞ = k2(M∞ − Mt)2

Weibull model Schott model

Note: Qt – amount of drug released or dissolved at time; Q0 – the initial amount of the drug in solution; K0 – zero order release constant; t – time; Q – amount of drug remaining at time t; K – first order release constant; KH – Higuchi dissolution constant; Mt./M∞ - fraction 1/3 – cube root of of drug released at time t; Q1/3 0 – cube root of the initial amount of drug; Qt remaining amount of drug at time t; F – fraction of drug release; k2 – second order rate constant.

I.E. Bordianu-Antochi et al. / Reactive and Functional Polymers 114 (2017) 118–126 Table 2 Kinetic study of CPF from the CH microspheres containing CPF. Mathematical model

pH = 1.2 buffer solution R2

PBS buffer solution R2

Zero order First order Higuchi model Hixson–Crowell Baker–Lonsdal Weibull model Schott model

0.654 0.996 0.807 0.767 0.778 0.954 0.993

0.516 0.977 0.682 0.627 0.668 0.858 0.984

and they correspond to a second–order diffusion kinetics, which means that the release mechanism is mainly controlled by the polymer chains relaxation and secondary by diffusion. 5. Conclusions

the polymer chains relaxations process. Also, in case of n value higher than 1 the mechanism is regarded as a Super Case II mechanism [49,50]. In the present paper, the data obtained after applying the mathematical models are resumed in Table 2 (corresponding to CH microspheres containing CPF) and Table 3 (corresponding to hybrid samples). Due to the fact that the Korsmeyer–Peppas model can be applied only for the first 60% of the drug released [47], for CH microspheres containing CPF this model was not applied, the burst release being rapid and high in both aqueous media. From Table 2 results that the highest values for R2 are recorded in case of Schott model, while from Table 3 one can observe that the highest values for R2 and n were obtained in case of Korsmeyer–Peppas model (applied only for the data obtained in PBS buffer solution) and Schott model. In this case, a value of 0.961 was calculated for R2, while 0.133 corresponds to n value, which indicates that the drug release mechanism in PBS solution corresponds to a Less Fickian mechanism. At longer times data there are some deviations of Fick's law and, in consequence, the Schott model was applied for data acquired at longer times in PBS buffer solution and for those obtained in SGF solution [48]. The mentioned model proved to fit well in both solutions due to the high values of R2, which resulted to be very close to unit (Table 3). Based on the applied mathematical models, a drug release profile of the samples could be outlined as follows: for CH microspheres containing CPF, a rapid burst release takes place in the first 30 min in both aqueous media, followed by a release mechanism which is mainly controlled by polymer chain relaxation and secondary by diffusion. The burst release of CPF from the hybrid samples in PBS buffer solution is slower than the one registered for CH microspheres containing CPF. In the next stages, the mechanism is controlled by diffusion in the first hours, while at longer times the relaxation rate of the polymer chains has a bigger influence as compared to diffusion process. To our knowledge, in literature there are only a few papers reporting this type of mechanism release and less for hybrid CH microspheres containing CPF or siloxane compounds [49,51–53]. In SGF solution, the hybrid samples present a faster burst release in the first 30 min. The main reason, beside the macroporous structure of the sample, is the presence of electrostatic repulsions between the polymer chains (after immersion in the dissolution medium) that allow the penetration of a higher amount of solvent. The data for the release kinetics come to sustain this hypothesis

Table 3 Kinetic study of CPF from the hybrid spheres. Mathematical model

pH = 1.2 buffer solution R2

PBS buffer solution R2

Zero order First order Higuchi model Hixson–Crowell Baker–Lonsdal Weibull model Schott model Korsmeyer–Peppas

0.505 0.631 0.667 0.587 0.616 0.861 0.997 R2 –

0.781 0.849 0.907 0.827 0.874 0.979 0.999 R2 0.961

n –

125

n 0.133

In the present study are reported the preparation and characterization of new hybrid materials based on CH microspheres and a siloxane compound intended for the delivery of CPF. The coacervation/precipitation method, which is a very simple and mild preparation method, was used to obtain CH microspheres, the last ones being incubated with CPF and subsequently coated with a siloxane compound. SEM data illustrated that the hybrid samples have a self–assembled macroporous structure which has proven to significantly influence the burst release and drug delivery process. The swelling capacity and the in vitro release mechanism were also investigated, the latest being dominated by diffusion and polymer chain relaxation. The combination between CH and the siloxane compound led to a matrix with enhanced properties in terms of resistance to dissolution, protection of the external layer and prolonged release of the drug. The increase of drug loading, improve of the burst release process and drug release percent are under study in order to extend the application field of such type of materials. References [1] V.B. Morris, C.P. Sharma, Update on Chitosan: A Non-viral Gene Delivery Vector, Smithers Rapra, Akron, 2012. [2] M.N.V. Ravi Kumar, A review of chitin and chitosan applications, React. Funct. Polym. 46 (2000) 1–27. [3] Y. Yang, S. Wang, Y. Wang, X. Wang, Q. Wang, M. Chen, Advances in self–assembled chitosan nanomaterials for drug delivery, Biotechnol. Adv. 32 (2014) 1301–1316. [4] M.C. Gutierrez, M. Jobbágy, M.L. Ferrer, F. del Monte, Enzymatic synthesis of amorphous calcium phosphate–chitosan nanocomposites and their processing into hierarchical structures, Chem. Mater. 20 (2008) 11–13. [5] K. Wong, G. Sun, X. Zhang, H. Dai, Y. Liu, C. He, K.W. Leong, PEI–g–chitosan, a novel gene delivery system with transfection efficiency comparable to polyethylenimine in vitro and after liver administration in vivo, Bioconjug. Chem. 17 (2006) 152–158. [6] N. Saranya, A. Moorthi, S. Saravanan, M.P. Devi, N. Selvamurugan, Chitosan and its derivatives for gene delivery, Int. J. Biol. Macromol. 48 (2011) 234–238. [7] Y. Liu, X. Cao, R. Hua, Y. Wang, Y. Liu, C. Pang, Y. Wang, Selective adsorption of uranyl ion on ion–imprinted chitosan/PVA cross–linked hydrogel, Hydrometallurgy 104 (2010) 150–155. [8] S. Spirka, G. Findeniga, A. Doliska, V.E. Reichel, N.L. Swansona, R. Kargl, V. Ribitscha, K. Stana-Kleinschek, Chitosan–silane sol–gel hybrid thin films with controllable layer thickness and morphology, Carbohydr. Polym. 93 (2013) 285–290. [9] Y.L. Liu, Y.H. Su, J.Y. Lai, In situ crosslinking of chitosan and formation of chitosan–silica hybrid membranes with using γ–glycidoxypropyltrimethoxysilane as a crosslinking agent, Polymer 45 (2004) 6831–6837. [10] X. He, M. Du, H. Li, T. Zhou, Removal of direct dyes from aqueous solution by oxidized starch cross–linked chitosan/silica hybrid membrane, Int. J. Biol. Macromol. 82 (2016) 174–181. [11] V.L. Goncalves, M.C.M. Laranjeira, V.T. Favere, R.C. Pedrosa, Effect of crosslinking agents on chitosan microspheres in controlled release of diclofenac, Polímeros 15 (2005) 6–12. [12] X. Xu, P. Dong, Y. Feng, F. Li, H. Yu, A simple strategy for preparation of spherical silica-supported porous chitosan matrix based on sol–gel reaction and simple treatment with ammonia solution, Anal. Methods 2 (2010) 546–551. [13] M. Vakili, M. Rafatullaha, B. Salamatiniab, M.H. Ibrahima, A.Z. Abdullah, Elimination of reactive blue 4 from aqueous solutions using 3–aminopropyl triethoxysilane modified chitosan beads, Carbohy. Polym. 132 (2015) 89–96. [14] P. Dhawade, R. Jagtap, Comparative study of physical and thermal properties of chitosan-silica hybrid coatings prepared by sol–gel method, Der. Chemica. Sinica. 3 (2012) 589–601. [15] C.M. Oliphant, G.M. Green, Quinolones: a comprehensive review, Am. Fam. Physician 65 (2002) 455–464. [16] A. Srinatha, J.K. Pandit, S. Singh, Ionic cross–linked chitosan beads for extended release of ciprofloxacin: in vitro characterization, Indian J. Pharm. Sci. 70 (2008) 16–21. [17] P.B. Kajjari, L.S. Manjeshwar, T.M. Aminabhavi, Novel interpenetrating polymer network hydrogel microspheres of chitosan and poly(acrylamide)–grafted–guar gum for controlled release of ciprofloxacin, Ind. Eng. Chem. Res. 50 (2011) 13280–13287. [18] A.F. Martins, P.V.A. Bueno, E.A.M.S. Almeida, F.H.A. Rodrigues, A.F. Rubira, E.C. Muniz, Characterization of N–trimethyl chitosan/alginate complexes and curcumin release, Int. J. Biol. Macromol. 57 (2013) 174–184. [19] A. Srinatha, J.K. Pandit, Alternate polyelectrolyte coating of chitosan beads for extending drug release, Drug Deliv. 15 (2008) 193–199. [20] J. Movaffagh, A. Ghodsi, B.S.F. Bazzaz, S.A.S. Tabassi, H.G. Azadi, The use of natural biopolymer of chitosan as biodegradable beads for local antibiotic delivery: release studies, Jundishapur. J. Nat. Pharm. Prod. 8 (2013) 27–33.

126

I.E. Bordianu-Antochi et al. / Reactive and Functional Polymers 114 (2017) 118–126

[21] K.S.V. Krishna Rao, K. Madhusudan Rao, P. Ramasubba Reddy, N. Sivagangi Reddy, Y. Shchipunov, C.S. Ha, Chitosan–poly(aminopropyl/phenylsilsesquioxane) hybrid nanocomposite membranes for antibacterial and drug delivery applications, Polym. Int. 64 (2014) 293–302. [22] M. Diaconu, A. Tache, S.A.M.V. Eremia, F. Gatea, S. Litescu, G.L. Radu, Structural characterization of chitosan coated silicon nanoparticles – a FT–IR approach, UPB Sci. Bull. 72 (2010) 115–122. [23] M.H. Hsiao, T.H. Tung, C.S. Hsiao, D.M. Liu, Nano–hybrid carboxymethyl–hexanoyl chitosan modified with (3–aminopropyl)triethoxysilane for camptothecin delivery, Carbohydr. Polym. 89 (2012) 632–639. [24] F. Quignard, R. Valentin, F. Di Renzo, Aerogel materials from marine polyssacharides, New J. Chem. 32 (2008) 1300–1310. [25] A. Madgulkar, M. Bhalekar, M. Swami, In vitro and in vivo studies on chitosan beads of losartan duolite AP143 complex, optimized by using statistical experimental design, J. Pharm. Sci. Technol. 10 (2009) 743–751. [26] N.E.A. El-Gamel, M.F. Hawash, M.A. Fahmey, Structure characterization and spectroscopic investigation of ciprofloxacin drug, J. Therm. Anal. Calorim. 108 (2012) 253–262. [27] R. Nithya, N. Meenakshi Sundaram, Biodegradation and cytotoxicity of ciprofloxacin-loaded hydroxyapatite-polycaprolactone nanocomposite film for sustainable bone implants, Int. J. Nanomedicine 10 (2015) 119–127. [28] S. Sahoo, C.K. Chakraborti, P.K. Behera, S.C. Mishra, FTIR and Raman spectroscopic investigations of norfloxacin/carbopol934 polymeric suspension, J. Young Pharm. 4 (2012) 138–145. [29] S.M.L. Silva, C.R.C. Braga, M.V.L. Fook, C.M.O. Raposo, L.H. Carvalho, E.L. Canedo, Application of infrared spectroscopy to analysis of chitosan/clay nanocomposites, in: T. Theophanides (Ed.), Infrared Spectroscopy–Materials Science, Engineering and Technology, InTech, Rijeka 2012, pp. 43–62. [30] S. Avaz, A. Taralp, Water soluble chitosan derivatives via the freeze concentration technique, Int. J. Chem. Molec. Nucl. Mater. Metall. Eng. 9 (2015) 334–337. [31] J. Kumirska, M. Czerwicka, Z. Kaczyński, A. Bychowska, K. Brzozowski, J. Thöming, P. Stepnowski, Application of spectroscopic methods for structural analysis of chitin and chitosan, Mar. Drugs 8 (2010) 1567–1636. [32] S. Sahoo, C.K. Chakraborti, S.C. Mishra, Qualitative analysis of controlled release ciprofloxacin/carbopol 934 mucoadhesive suspension, J. Adv. Pharm. Technol. Res. 2 (2011) 195–204. [33] J.E. dos Santos, E.R. Dockal, E.T.G. Cavalheiro, Synthesis and characterization of Schiff base derivatives from chitosan and salicylaldehyde derivates, Carbohydr. Polym. 60 (2005) 277–282. [34] A. Ghorbani-Choghamarani, P. Moradi, B. Tahmasbi, Ni-S-methylisothiourea complex supported on boehmite nanoparticles and its application in the synthesis of 5-substituted tetrazoles, RSC Adv. 6 (2016) 56638–56646. [35] S.H. Jun, E.J. Lee, S.W. Yook, H.E. Kim, H.W. Kim, Y.H. Koh, A bioactive coating of a silica xerogel/chitosan hybrid on titanium by a room temperature sol–gel process, Acta Biomater. 6 (2010) 302–307. [36] R. Zolfaghar, M. Shahbaei, Bionanocomposites based on alginate and chitosan/layered double hydroxide with ciprofloxacin drug: investigation of structure and controlled release properties, Polym. Compos. 36 (2014) 1819–1825.

[37] L. Yu, D. Wang, W. Hu, H. Li, M. Tang, Study on the preparation and adsorption thermodynamics of chitosan microsphere resins, Front. Chem. Chin. 4 (2009) 160–167. [38] L. Liao, X. Li, Y. Cui, Study on the swelling, shrinking and bending behavior of electric sensitive poly (2–acrylamido–2–methylpropane sulfonic acid) hydrogel, Mod. Appl. Sci. 7 (2009) 115–120. [39] Z. Ozbas, G. Gurdag, Swelling kinetics, mechanical properties, and release characteristics of chitosan semi–IPN hydrogels, J. Appl. Polym. Sci. 132 (2015) 41886–41897. [40] K. Varaprasad, K. Vimala, S. Ravindra, N. Narayana Reddy, G. Siva Mohana, K. Mohana Raju Reddy, Biodegradable chitosan hydrogels for in vitro drug release studies of 5–flurouracil an anticancer drug, J. Polym. Environ. 20 (2012) 573–582. [41] X. Huang, C.S. Brazel, On the importance and mechanisms of burst release in matrixcontrolled drug delivery systems, J. Control. Release 73 (2001) 121–136. [42] A. Thakur, S. Monga, R.K. Wanchoo, Sorption and drug release studies from semi-interpenetrating polymer networks of chitosan and xanthan gum, Chem. Biochem. Eng. Q. 28 (2014) 105–115. [43] L. Zhao, B. Zhu, Y. Jia, W. Hou, C. Su, Preparation of biocompatible carboxymethyl chitosan nanoparticles for delivery of antibiotic drug, Biomed. Res. Int. (2013)http://dx.doi.org/10.1155/2013/236469. [44] Z. Orhan, E. Cevher, L. Mülazimoglu, D. Gürcan, M. Alper, A. Araman, Y. Özsoy, The preparation of ciprofloxacin hydrochloride–loaded chitosan and pectin microspheres, J. Bone Joint Surg. 88-B (2006) 270–275. [45] S. Minko, Stimuli responsive fine particles, in: E. Matijević (Ed.), Fine Particles in Medicine and Pharmacy, Springer, New York 2012, pp. 283–902. [46] J.M. Souza, A.L. Caldas, S.D. Tohidi, J. Molina, A.P. Souto, R. Fangueiro, A. Zille, Properties and controlled release of chitosan microencapsulated limonene oil, Rev. Bras. Farm. 24 (2014) 691–698. [47] M.H. Shoaib, S.A.S. Siddiqi, R.I. Yousuf, K. Zaheer, M. Hanif, S. Rehana, S. Jabeen, Development and evaluation of hydrophilic colloid matrix of famotidine tablets, J. Pharm. Sci. Technol. 11 (2010) 708–718. [48] S.K. Bajpai, F.F. Shah, M. Bajpai, Dynamic release of gentamicin sulfate (GS) from alginate dialdehyde (AD)–crosslinked casein (CAS) films for antimicrobial applications, Des. Monomers Polym. 20 (2016) 18–32. [49] Synthesis, swelling properties, and network structure of new stimuli-responsive hydrogels [50] J. Wang, W. Wu, Z. Lin, Kinetics and thermodynamics of the water sorption of 2hydroxyethyl methacrylate/styrene copolymer hydrogels, J. Appl. Polym. Sci. 109 (2008) 3018–3023. [51] M. Gierszewska-Drużyńska, J. Ostrowska-Czubenko, Mechanism of water diffusion into noncrosslinked and ionically crosslinked chitosan membranes, PCACD, 17, 2012, pp. 59–66. [52] F. Ganji, S. Vasheghani-Farahani, E. Vasheghani-Farahani, Theoretical description of hydrogel swelling: a review, Iran. Polym. J. 19 (2010) 375–398. [53] C. Wang, B. Yu, B. Knudsen, J. Harmon, F. Moussy, Y. Moussy, Synthesis and performance of novel hydrogels coatings for implantable glucose sensors, Biomacromolecules 9 (2008) 561–567.