Oral administration is one of the most preferred drug delivery ways because of convenience and no cross-infection . Many different kind of polymeric systems are proposed as drug carriers for oral delivery with the aim of site-specific delivery such that the drug is released at controlled rate and at the desired time [2-5]. Stimuli-responsive polymers are a group of materials that swell considerably in external stimuli and demonstrate extraordinary capacity [5, 6]. Being insoluble, these networks can retain a large amount of water, which contribute to their good blood compatibility, beside maintain a certain degree of structural elasticity and integrity [7, 8]. These materials exhibit phase transitions in response to change in pH, ionic strength, temperature, light, magnetic or electric field and etc. . The most important stimuli used for drug delivery are pH and temperature [10, 11]. The temperature has to be altered externally within narrow limits in most cases except maybe hyperthermia therapy, but the pH changes extensively within the body [12, 13]. The obvious pH changes in the gastro-intestine track from acidic in the stomach (pH 1-3) to basic in the intestine (pH 5-8) make the pH a suitable stimulus and considered for oral delivery of any kind of drug [14-16].
Nowadays, development of nano-structured materials in the field of biology and medicine caused interesting applications. Iron oxide nanoparticles (Fe3O4 and Fe2O3), in particular, have received special attention for their strong magnetic properties and low toxicity [17, 18].
Magnetic drug targeting has been used to improve localized drug delivery by application of an external magnetic field.
For a given period of time, the magnetic field retained until receiving very high concentrations of the therapeutic agents near the target tissue without any toxic effects to surrounding tissues or the whole body . By combination, magnetic nanoparticles (MNPs) covered with a layer of biocompatible polymeric shell have been reported to be effective drug carriers [20, 21]. The polymer coatings have found to reduce aggregation problem of uncoated nanoparticles and lower toxicity . Furthermore, the therapeutic drug concentration is maintained for sustained periods of time in polymeric network and prevents drug degradation. In addition, control of release kinetics of the drug from MNPs can be easily modulated by altering formulation parameters such as polymer molecular weight and composition, drug: polymer ratio, etc .
It is reported that there is a relationship between the release rate of the drug and the degree of interaction between the polymer and the drug [24, 25].
Drug-polymer interaction may be physical through hydrogen bonding or chemical through the formation of insoluble complexes between them. Otherwise, changing in the interactions occurs when the carrier comes in contact with the release medium and affects the rate and extent of drug release. However, it is remains difficult to predict precise drug release profile.
The three main mechanisms of releasing from the polymeric magnetic MNPs are diffusion, degradation, and swelling followed by diffusion .
Diffusion occurs due to concentration gradient between release medium and carrier, thus migrate out of the drug. Degradation occurs when drug molecules, which trapped by the polymer, release due to hydrolyzes of polymer chains. Finally, in swelling controlled release, when the polymer placed in the release fluid, swells to increase porosity, enabling the drug to diffuse from network.
pH-responsive polymers can swell when the pH approaches the dissociation constant (pKa) of the ionic monomer incorporated within the network . Many magnetic polymeric core-shell nano-carriers were used for drug delivery [28-31].
PVP based polymers have been used as drug delivery systems. The choice of PVP was due to its long-standing and safe record in pharmaceutical applications. Additionally, it has been reported that PVP enhance drug circulatory time in plasma [32, 33].
Presence of functional groups like -COOH make the polymer more suitable to bind with the drugs, methacrylic acid and acrylic acid are suitable monomers to form cationic polymer grafted nonionic PVP. This co-polymer will contain acidic carboxyl groups that partially dissociate in water, producing a flexible coil structure.
The designed matrix dosage form can absorb water molecules, which is accompanied by polymer chain relaxation and causes the drug is entrapped in the polymeric network and released over a long period of time.
In this study, iron oxide nanoparticles were synthesized by chemical co-precipitation technique and primary coated with 3-trimethoxysi- lylpropylamine (APTS).
The secondary coating was performed using pH-responsive poly (MAA-co-AAc)-grafted PVP. The structure and morphological characterization was performed out by TEM and XRD. The chemically covalent interactions were investigated by FT-IR and elemental analysis and TGA, and DSC studied thermal stability. This carrier was loaded with two antibiotics, ciprofloxacin and ofloxacin, as model drugs (Fig. 1).
Fig. 1. Chemical structure of (a) ciprofloxacin and (b) ofloxacin
The loading kinetics and in-vitro drug release profile of the drugs were studied in three different pH (1.2, 5.5, and 7.2) buffered solutions at 37ºC. Different parameters affecting on the release of the drugs such as the amount of MNPs, initiator, cross-linker, and MAA content were evaluated.
The swelling behavior and drug release kinetics and mechanism were also studied.
The chemicals were listed below were purchased from Merck Company and used without further purification. Ferrous chloride hexahydrate (FeCl3.6H2O), ferric chloride tetrahydrate (FeCl2.4H2O), aqueous ammonia (25%), 3-trimethoxysilyl propylamine (APTS), methanol, acetic acid, sodium acetate trihydrate (CH3COONa.3H2O), sodium hydroxide, potassium hydrogen phosphate, Glycerol (about 87%), hydrochloric acid, potassium chloride, acrylic acid (AAc), methacrylic acid (MAA), and ethylene glycol dimethacrylate (EGDMA, 98%). Ciprofloxacin and ofloxacin were received as a gift from Faculty of Pharmacy, Tehran University. Deionized water was used throughout the experiment. 2, 2´-Azobis isobutyronitrile (AIBN, 98%) was purchased from ACROSS and polyvinylpyrrolidone (PVP, Mw40 000) was purchased from Rahavard Tamin chemical Co. (Iran, Saveh).
A CM120 and anEM2085 Phillips transmission electron microscopes (Netherlands) with an accelerating voltage of 100 kV were used for characterization of particle size and morphology of APTS-Fe3O4 nanoparticles and polymer coated MNPs respectively. Phase characterization was performed using Phillips PW-1800 X-Ray diffraction. Characterization of Fe3O4 nanoparticles, APTS coated Fe3O4 nanoparticles, polymer coated MNPs, and drugs loaded magnetic nano-carrier were carried out by Perkin Elmer FT-IR spectrometer, Spectrum one Bv5.3.0.
Elemental analysis was performed onto the APTS-coated Fe3O4 nanoparticles, polymer coated MNPs, and drugs loaded magnetic nano-carrier\using Perkin Elmer 2400 SERESE II. The thermal behavior of poly (MAA-co-AAc)-grafted PVP and poly (MAA-co-AAc)-grafted PVP coated MNPs were characterized by differential scanning calorimetry and thermo gravimetric analysis, STA-1500, Rheometric Scientific. UV-Vis spectra of released drugs and other solutions were carried out by UV-240 Shimadzu UV-Vis spectrophotometer. A Metrohm 827 pH meter was used for measuring and adjusting the pH of the solutions. A water bath Memmert WB10 was used to precise control of the temperature (37 ± 0.1 oC).
Coating of Fe3O4 nanoparticles with APTS
Core-shell strategy was applied to synthesis the nanoparticles. First, the APTS coating was performed as follow.
The Fe3O4 nanoparticles were prepared via chemical co-precipitation method with some modification .Briefly, 11.6 g ferrous chloride hexahydrate (FeCl3.6H2O) and 4.3 g ferric chloride tetrahydrate (FeCl2.4H2O) were dissolved in 250 mL deionized water under nitrogen atmosphere with vigorous stirring as the temperature rises to 85 oC. Then 40 mL of 25% aqueous ammonia was added to the solution.
The color of the solution turned immediately from orange to black. Then the solution left to reach ambient temperature and washed with deionized water and 0.02 M sodium chloride to neutralize. Then, the magnetic suspension was placed in a 250 mL round bottom flask and allows participates to settle. The supernatant was removed, and an aqueous solution of 80 mL 10 % (v/v) APTS, followed by 60 mL glycerol was added.
The mixture was then stirred and heated at 90 oC for 2 h under nitrogen atmosphere. After cooling, the suspension was washed sequentially with 200 mL deionized water(three times), 100 mL methanol (two times) and 200 mL deionized water (three times). The MNPs dried to powders at 50 oC in oven for further coating with the polymer.
Coating of APTS- MNPs with poly (MAA-co-AAc)-grafted PVP
The free radical polymerization was carried out in 50 mL conical flask. 1 mL of 0.37% PVPin methanol and 1mL 0.49 mM of AAc were dissolved in 12 mL methanol. Variable amount of MAA (0.7-3 mL, 0.39 mM) was added simultaneously. Variable amount of EGDMA (0.7-3 mL, 0.39 mM) as cross-linker, variable amount of AIBN (0.3-1.2 mmol) as initiator and variable amount of MNPs (0-100 mg) were added to the reaction mixture after addition of monomers. Then the mixture was bubbled with N2 gas to remove dissolved oxygen and continuously stirred under the nitrogen atmosphere for 15 min. The cross-linking reaction was completed within six hours at 60 ºC in water bath. After cooling, the polymeric particles were separated from the polymerization medium and washed three times with 200 mL methanol and five times with 250 mL deionized water. The UV-Vis spectra of the supernatant were compared with the reagents. Elimination of all reagent peaks in the supernatant confirms the removal of unreacted reagents. The collected nanoparticles were dried in oven at 50 ºC to obtain fine powder and stored in desiccators before use. Polymer with 20% (v/v) of MAA and EGDMA and 0.6 mmol of AIBN was used as plain and that with 20% (v/v) of MAA, 15% (v/v) of EGDMA, 0.6 mmol of AIBN and 10 mg of MNPs was used as the nano-carrier for further study of swelling and loading behaviors.
The size and morphology of APTS-Fe3O4 nanoparticles and polymer coated MNPs was characterized using TEM images. The images were obtained by placing one drop of each sample on a carbon plate. XRD technique was used to confirm the synthesis of APTS coated Fe3O4 nanoparticles using Cu-Ká radiation wavelength 1.540598 Aº, matching peak position and relative intensities to the Joint Committee on Powder Diffraction Standards (JCPDS card, file No. 79-0418). To confirm the structure of APTS-Fe3O4 nanoparticles, polymer coated MNPs, plain polymer, ciprofloxacin-loaded and ofloxacin-loaded magnetic nanoparticles, FT-IR spectra were recorded in transmission mode in the 400-4000 cm-1 region for sample dispersed in KBr pellets. Stepwise increase in organic compartments (carbon, hydrogen, oxygen and nitrogen), which observed by elemental analysis, was used to confirm the coatings of MNPs and loading of the drugs. The volume-phase transition behavior of the plain polymer and polymer coated MNPs were investigated by TGA and DSC using STA techniques. It is capable of providing direct weight measurements by time-programmed thermogravimetric analysis (TGA) as well as heat consumption release by differential scanning calorimetry (DSC) to detect thermal events that do not involve mass change. The method combines the benefits of thermal analysis and differential scanning calorimetry techniques into a single experiment. Scans were run from ambient temperature to 800 ºC at a heating rate of 10 ºC min-1. Plain polymer and polymer coated MNPs were dried at 50 ºC for 24 h and homogenized into fine powder before used.
To determine the pH-dependent swelling properties of the plain polymer, weighed dry samples were immersed in buffered solutions with pH of 1.2-9.0. When the swelling was equilibrated, the swollen polymer separated from the solution, their surface water was removed with tissue paper and weighed. This procedure was repeated for each pH value. The following equation was used to determine the swelling ratio:
Where, Wt is the weight of the swollen polymer and W0 is the weight of the polymer before swelling experiments. Each swelling process was carried out for approximately 148 h. Then, the samples were removed from the solution, weighted and dried for the FT-IR studies (not mentioned here).
Drug content and encapsulation efficiency
A 0.5 g of dried polymer and polymer coated MNPs were weighed and various amount of ciprofloxacin and ofloxacin solutions were added to make different sorbent to drug weight ratio. The solutions were stirred well to mix homogeneously and left to equilibrate for 24 h at room temperature. Unbounded ciprofloxacin and ofloxacin were determined by UV-Vis spectrophotometry at 279 and 298 nm. The supernatant from unloaded polymer and polymer coated MNPs were used as blank. As necessary, appropriate dilutions were performed to ensure that the absorbance of the drugs was in the range of Beer’s law. The % drug content and encapsulation efficiency were calculated using the following equations:
Where, WDrug loaded is the weight of ciprofloxacin or ofloxacin encapsulated in the carrier. Wcaeeier is the weight of each carrier andTotal drug is the total weight of free ciprofloxacin or ofloxacin. The drug-loaded carriers were washed five times with deionized water to remove unloaded drugs for further release studies.
In-vitro drug release
The in-vitro release experiments were performed in three buffered solutions at 37 ºC. The pH values were simulated to gastric fluid (pH 1.2), duodenum and intestine fluids (pH 5.5 and 7.2). Ciprofloxacin and ofloxacin loaded carriers from loading procedure were immersed into 25 mL of each buffered solution.
At certain time intervals, 6 mL of the supernatant was removed and the same volume of buffer solution was replaced. The amount of ciprofloxacin and ofloxacin released from the nano-carries were determined by spectrophotometry. The effect of different parameters on the drug release was also investigated. The data mentioned hereis the average of three replicate experiments.
Kinetic modeling of drug release
Consideration of data obtained in release experiments was easier when mathematical formulas, which can express the release results as a function of carrier characteristics, are used. Among various kinetics models, the release profile of ciprofloxacin and ofloxacin were fitted to zero-order, first-order, Higuchi, Hixson-Crowell, Korsmeyer-Peppas, and Beaker-Lonsdale models. The zero-order model describes the systems were drug release is independent of its concentration and can be represented by the equation:
Qt = Q0 + K0t
Where Qt is the amount of drug released at time t, Q0 is the initial amount of the drug in solution (Q0=0 in most times) and K0 is the rate constant. The first-order model describes the release of drug from systems where release rate is concentration dependent and can be expressed by the equation:
Where C0 is the initial concentration of drug and K1 is the first-order rate constant. This relationship can be used to describe release kinetics from systems containing water-soluble drugs in porous matrix. Higuchi model is the first example of a mathematical model aimed to describe the release of drugs from insoluble matrix. The simplified Higuchi model can be expressed as:
Where KH is the Higuchi release constant reflecting the design variables of the system.
Korsmeyer-Peppas derived a simple equation, which described drug release from polymeric system which can represent as:
Where Mt/M¥ is a fraction of drug released at time t, KKC is the release rate constant and n is the release exponent. To find out the mechanism of drug release, first 60% drug release data were fitted in Korsmeyer-Peppas model. The n value is used to characterize different release mechanism for different shaped matrix as shown in Table 1.
Table 1. Release exponent value and related mechanism
The Baker-Lonsdale model was developed from the Higuchi model and describes the drug release from spherical matrixes according to the equation:
Where, K is the release constant corresponding to the slope.
RESULTS AND DISCUSSION
The TEM images of APTS-Fe3O4 nanoparticles and polymer coated MNPs are shown in Fig. 2. APTS-Fe3O4 nanoparticles image (Fig. 2a) shows good size distribution with average diameter of 8.6 ±1.6 nm and The TEM image of polymer coated MNPs (Fig. 2b) show that the polymer coated MNPs have spherical shape yet with average size of 19 ±2.1 nm. X-ray diffraction pattern of APTS coated Fe3O4 nanoparticles were shown in Fig. 3 and the indexing results are summarized in table 2. The pattern indicates that the nanoparticles have a cubic structure and modification did not change the crystal structure of the MNPs.
Table 2. Comparison of APTS-MNPs peak position (2q) values and diffracting plane index with naked Fe3O4 nanoparticles
Fig. 2. TEM images of (a) APTS- Fe3O4 NPs and (b) polymer coated MNPs
Fig. 3. The XRD pattern of prepared APTS coated MNPs
FT-IR is an appropriate technique to study the chemical interaction or adsorption of the reagents. The FT-IR spectra of APTS coated Fe3O4 nanoparticles, polymer coated MNPs, plain polymer, and ciprofloxacin and oflaxacin loaded magnetic nano-carrier are shown in Fig. 4. The absorption band of Fe-O-Si, which indicates the bonding of APTS on the magnetic nanoparticles, cannot be seen in the FT-IR spectrum because it appears at around 585 cm-1and therefore overlaps with Fe-O vibration band of the nanoparticles. However, two bands at 1125, 1039 and 979 cm-1 assigned to Si-O-H and Si-O-Si vibration bond. Similarity in the spectra of the plain polymer and polymer coated MNPs (Fig. 4b and c)beside the presence of characteristic peak of Fe3O4 nanoparticles absorption peak along with its absence in plain polymer confirm the coating of the polymer onto the APTS coated nanoparticles. Loading of each drug was confirmed with addition of absorption peaks in 1625 cm-1 (Fig. 4d) and 1622 cm-1 (Fig. 4e) which is assigned to C=C ring stretching of ciprofloxacin and ofloxacin. Fig 5 shows the TGA-DSC plots of the plain polymer and polymer coated MNPs. From Fig. 5a, a weight loss with endothermic peak at about 60 oC may attribute to the physically adsorbed water. The three exothermic peaks at 345, 388 and 445 may attributed to the degradation of PVP and poly carboxylic acid network. As can be seen in Fig. 5b, polymer coated MNPs show good thermal stability until 300oC. Then, the degradation will occur. Furthermore, elemental analysis was performed onto APTS-Fe3O4NPs, polymer coated MNPs and ciprofloxacin-loaded and ofloxacin-loaded magnetic nano-carriers.
The results show an increase in the amount of organic compounds (hydrogen, carbon and nitrogen), which confirm the functionalization of MNPs with APTS, coating with the polymer, and loading of the drugs.
Fig. 4. FT-IR spectra of (a) APTS-Fe3O4 NPs,(b) polymer coated MNPs, (c) plain polymer, (d)ciprofloxacin loaded nano-carrier and (e) ofloxacin loaded nano-carrier
Fig. 5. TGA-DSC plots of (a) plain polymer and (b) polymer coated MNPs
The pH-sensitive swelling studies of poly (MAA-co-AAc)-grafted PVP in different buffer solutions were shown in Fig. 6.
Fig. 6. The swelling behavior of poly (MAA-co-AAc)-grafted PVP in different pH
As can be seen from the Fig, increase in the pH of buffer solution was caused the increase in swelling ratio of the polymer up to pH 8.5 to form a gel media. Both PMAA and PAAc monomers are weak acids with pKa 7.0 and 6.6 respectively. In the pHs lower than the pKa, high concentration of the H+ ions cause most carboxylic groups were in the form of protonated non-ionic (-COOH) groups. Hydrogen bonding constructed by the carboxylic acid groups together and with –C=O groups of PVP led to the stronger interaction between polymer chains and decrease the swelling (gel collapsing).
As the pH increased above the pKa, ionization of the carboxylic groups occurs and the electrostatic repulsion between the same charged groups are favor of the swelling. Other authors reported similar observation previously [34, 36].
Drug content and encapsulation efficiency
The drug content and encapsulation efficiency of plain polymer and polymer coated MNPs (carriers) were determined by varying the weight ratio of the carrier to drugs. Table 3 summarizes the loading characteristics of different weight ratio for plain polymer and polymer coated MNPs. The long entrapment time (24 h) would allow a good chance for interaction between the polymer and the drug. For plain polymer, when the weight ratio of the carrier to drug was 2.5:1, the drug content was about 37.6-38.8 % with encapsulation efficiency of 94-97 %. With increasing the carrier to drug ratio (i.e. the amount of the drug per carrier mass unit decreased), a decrease in encapsulation efficiency occurs. The same behavior was observed for polymer coated MNPs with slightly decreased in drug content to about 35.6-36.6 % (encapsulation efficiency of 89-91 %). This may occurs due to lower polymer content in polymer coated MNPs carrier with the same weight as plain polymer which decreases the overall interactions of the drug with the carrier.
Table 3. Loading characteristics of plain polymer and polymer coated MNPs for ciprofloxacin and ofloxacin
Drug release studies
Effect of pH
The drug release profiles of ciprofloxacin and ofloxacin for polymer coated MNPs were shown in Fig. 7. As could be seen, the drug release at pH 1.2 is much faster than at pH 5.5 and 7.2. As mentioned before, when the pH is increased from 1.2 to 7.2 (above the pKa of The PMAA and PAAc), an increases in dissociation of polymer carboxylic groups occurs, which facilitate the interaction of anionic carboxyl groups of the polymer with amine group of the drugs and probably interaction of carboxyl group of the drugs with the amine groups of APTS. Furthermore, when the interactions between the polymer and the drugs occur, an insoluble complex is formed, which further retards the release. In addition, at pH values above the pKa’s, the carboxylic acid groups of the polymer ionized and repulsion of the chains cause expansion of the polymer and form a gel that slows drug release rate. The sudden dip in release rate of ciprofloxacin and ofloxacin in pH 5.5 and 7.2 may be explained in part by these phenomena. The effect of decrease in drug release with increasing pH, i.e. increase in the interaction of anionic polymers with cationic drug which decrease the amount of drug release, were observed by many authors [37-39].On the other hand, the swelling of the polymer increases with increasing the pH value, thus, a little increase in atorvastatin release % was observed in pH 7.2 with respect to pH 5.5 may explained by increasing diffusional pass due to expanding the polymer.
Fig. 7. The release profile of (a) ciprofloxacin and (b) ofloxacin in different pH
Effect of the amount of MNPs
The effect of amount of MNPs on the ciprofloxacin and ofloxacin release was studied by varying the amount of MNPs incorporated in the carrier in the range of 10-100 mg.
Experimental results indicate that the amount of the drugs released from the carrier decreased with increasing the amount of MNPs. This observation can be explained by decreasing in polymer content of the carrier with increasing in the MNPs amount, which decreases total interactions of the drugs due to lower pore density. Thus decreased the amount of drug released.
Effect of methacrylic acid content
Variation in the amount of MAA can alter the amount of carboxylic acid groups in the polymeric shell and altering in chemical structure can affects on the release of the drugs. Thus the amount of MAA varies in the range of 3-14 % (v/v), in presence of 10 mg Fe3O4 nanoparticles. As could be seen from Fig.8, increase in MAA content up to 10 % (v/v) raises the drug released from the carrier because of increase in chain porosity, therefore hydrodynamic free volumes increase which increases the drug release. A drop in the release after this point could be explained, as the polymeric chains on the MNPs were largely crowded and the release of drug molecules not allowed.