Over the past few years, particular interest has been paid towards the fabrication of new drug delivery systems (DDS) to specifically deliver the therapeutic agents to target sites [1-2]. This technology offers great advantages over conventional DDS such as low toxicity, wide therapeutic window, minimized side effects, ideal drug efficacy and releasing drugs in a sustained rate . Magnetic targeting can be achieved through the introduction of magnetic nanoparticles into the drug-carrier matrices . Magnetic nanoparticles have been of considerable interest for their practical biomedical applications. Particularly, ultrafine size and highly crystalline magnetic iron oxide nanoparticles (MNPs) is garnering substantial attention because of possessing superior characteristics like superparamagnetism, high surface-to-volume ratio, high field saturation and biocompatibility [5-6]. However, the main challenge is self-aggregation of these particles because of their high surface energy and strong magnetic dipole–dipole interactions. Most importantly, for biomedical applications, it is of fundamental importance modifying the surface of nanoparticles with water-soluble and specific functionalized groups to enable conjugation of the drug . Therefore developing well-defined strategies for functionalization of MNPs with well-considered nanostructures is in great demand. Of all the available materials, non-toxic polysaccharide based hydrogels such as chitosan, alginate, Carrageenan, starch, gelatin and carboxy methyl cellulose is increasingly being used as one of the most significant matrices in new biomedicine applications [8-10]. Among all the polysaccharides, carboxymethyl cellulose (CMC) is the most widely used and easily available cellulose ether today. It is biocompatible, biodegradable, hydrophilic, non-toxic, and very cheap [11-12]. Owing to a number of sodium carboxymethyl (CH2COONa) and hydroxyl (-OH-) groups promote water solubility and high chemical stability . It has been used in various drug delivery applications especially as bioadhesive material, for nasal and buccal drug delivery systems . A crucial perspective for application of hydrogels as drug delivery systems is the degree of swelling. However, a single problem associated with use of conventional hydrogels in drug delivery purposes is weak and fragile architecture because of the large proportion of water penetrated in their structure . To address this problem and enhance stability, one strategy is based on the use of nanoclays . In order to modify drug release, a fascinating approach is to use polymer/layered silicate nanocomposites. These hybrid materials carryout specific functions such as modifying drug delivery patterns, improving drug dissolution and increasing drug stability . Among layered silicate materials, considerable attention has been drawn towards montmorillonite (MT) nanoclay because of possessing substantial properties such as large specific surface area, rheology, chemical inertness, bioadhesion, high cation exchange capacity and good absorbance ability . Additionally, MT has also been found to be non-toxic by histopathological, hematological, and biochemical investigations in rat models . These inherent advantages make MT-Clay mineral to be used successfully as vehicles for delivery of drug molecules. Some shiny examples include intercalation of donepezil in montmorillonite for Alzheimer disease . Intercalation of 5- fluorouracil with MT as a drug vector  and encapsulation of paclitaxel in Poly (D, L-lactideco-glycolide)/ MT for breast cancer chemotherapy . The novelty of present study was development of dual magnetic/pH sensitive hydrogel composed of carboxymethyl cellulose–g-acrylamide in the presence of montmorillonite nanoclaywith the ultimate goal of the medically applicable device for colon specific drug delivery through oral administration.
MATERIALS AND METHODS
Chemicals and Reagents
Sodium carboxymethyl cellulose (NaCMC) and diclofenac sodium were purchased form Sigma–Aldrich (St. Louis, USA). Natural sodium-montmorillonite (sodium Cloisite, Na-MT) as a clay with cation exchange capacity of 92 meq/100 g of clay was provided by Southern Clay Products (USA). Acrylamide was purchased from Nalco Chemical Co. (Netherlands) and used after recrystallization from acetone. N, N-methylenebisacrylamide and ammonium persulfate were purchased from Merk (Germany). All other chemicals were analytical grade and used without any purification.
Synthesis of magnetic montmorillonite (mMT)
The magnetic montmorillonite (mMT) was manufactured according to our previous work . 3g of sodium montmorillonite was dispersed in 200 mL distilled water and stirred for overnight. The dispersed clay was then sonicated for 20 min under the induced frequency of 60 kHz. Afterward, the certain amount of Fe3+/Fe2+ with a molar ratio of 1 : 2 were dissolved in distilled water with vigorous mechanical stirring for 90 min. Two solutions with the ratio reported by Wu et al were mixed and allowed to stir for 30 min .Then, the ammonia (3M) was added dropwise into the above solutions and allowed to reach the pH of 11.The prepared mMT was stirred at 70oC for 2h and consequently cooled to ambient temperature. The solution was centrifuged and washed with distilled water several times. Finally, the mMT was separated by a magnet and the volume of solution was adjusted to 80 mL.
Preparation of PAm grafted CMC intercalated mMT (CMC-g-Am/mMT)
CMC-g-Am/mMT hydrogels were synthesized via grafting of Am onto CMC using MBA as a crosslinking agent, KPS, as a free radical initiator, and mMT nanoclay as nano-filler. In the undertaken protocol, 1 g of CMC was poured in distilled water and was stirred atambient temperature for 90 min. 3 g of Am monomer and 0.10 g of MBA (dissolved in 2 mL of distilled water) were added into the above solution. The mixture was stirred for 1 h and subsequently 0.10g of the APS was added to the reactor while stirring. After that, mMT solution of different volumes (0, 10 and 15 mL) was added to the reaction media and sonicated for 10 min. The resultant were collected and washed with excess water, dried at ambient temperature for constant weight. The as-prepared resultants were coded as CMC-g-Am, CMC-g-Am/mMT1 and CMC-g-Am/mMT2.
Water uptake measurements
To study water absorbency (WA), two different sets of experiments were performed which include dynamic swelling measurement at simulated intestinal fluid (SIF) (phosphate buffer solution (PBS, pH 7.4): 8g NaCl, 0.2g KCl, 1.44 Na2HPO4, 0.24 KH2PO4, 1L H2O) and simulated gastric fluid (SGF) (hydrochloric acid buffer solution (HBS, pH 1.2): 0.15M HCl, 0.05M KCl) and kinetic swelling measurement in distilled water. For swelling measurements, the dry hydrogels (~0.2g) were placed in media and kept at a constant temperature, 20±0.5oC. Swollen samples were periodically removed, weighted and the water absorbency was measured according to Eq. 1:
Where Ws and Wd are the weights of the samples swollen in buffers and in a dry state, respectively. The all swelling data were repeated three times and the mean values were shown in graphs (mean ± S.D. n=3).
Diclofenac sodium (DS) was used as a model drug. For loading DS molecules, 100mg of hydrogels were left to soak in 3mL of an aqueous solution of DS (500mg/L) at room temperature for 48h. After this time, samples were withdrawn from drug solution and washed with distilled water to remove the adsorbed drug on the surface of hydrogels, and the liquid was taken to be analyzed by UV−vis absorption spectroscopy by monitoring the absorbance changes at λmax=276nm. The amount of DS loaded by the hydrogels was calculated by the following Eq. 2:
Where, C0 is the initial concentration of DS in the solution (mg/L) and Ct is the concentration of DS remained in the media (mg/L)
In vitro drug release
The release profile of Diclofenac sodium (DS) loaded hydrogels was evaluated at different pHs (1.2 and 7.4) and magnetic field (100 and 300 G). In each drug release experiment, certain amount of drug loaded hydrogels was sealed into 20 mL buffered solutions with pH 1.2 and 7.4.The release medium (5mL) was withdrawn at predetermined time intervals and the amount of released drug was quantified spectrophotometrically at λmax=276 nm. After each measurement, the withdrawn medium was replaced by 5 mL of fresh buffered solution. The cumulative release (R, %) of drug from samples was calculated according to Eq. 3:
Where Mt and M∞ are the drug released at time t and initial loaded drug content, respectively. The all releasing data were repeated 3 times and the mean values were shown in graphs (means±SD. n=3). An external alternative magnetic field (AMF) apparatus (Electromagnet Type E, Newport Instruments, England) was also employed to monitor the magnetically drug release in phosphate buffer pH=7.4 . Magnetic field frequency of 350 kHz and the strength of magnetic fields (100 and 300G) were set by adjusting the voltage of the equipment. The hydrogels were immersed in 10 mL of buffered solution with pH 7.4 and kept between magnetic poles for 180 min. The supernatant was used to calculate the amount of released drug during 180 min.
The surface morphology ofhydrogels was observed by field-emission scanning electron microscope (FESEM;NanoSEM 630, NOVA, USA) equipped with an energy-dispersive X-ray analysis system (EDXA).X-ray diffraction (XRD) pattern of thehydrogels was recorded on a Siemens D-500 X-ray diffractometer (ESCALAB220i-XL, VG Scientific, UK) with Cu Ka radiation (30kV, 30mA)in the 2θ range from 2 to 80 °. The magnetic properties of the products were measured on a Lake Shore 7400 vibrating samplemagnetometer (VSM, Model 7400, Lakeshore Company, USA).Transmission electron microscopy(TEM) images were takenwith a Zeiss EM 900TEM (Zeiss SMT, Oberkochen, Germany) at 80 kV.FT-IR Spectrawere recorded between 400–4000cm-1 bya Bruker 113V FT-IR spectrometer. (USA). Ultraviolet–visible (UV-Vis) absorption spectra were measured on an UV-VIS-NIR spectrophotometer (Shimadzu UV-Vis 3600, Japan).
RESULTS AND DISCUSSION
Synthesis and characterization
The magnetic nanocomposite hydrogels (CMC-g-Am/mMT) were prepared by the methodology reported by Mohapatra. Schematic route for the synthesis of CMC-g-Am/mMT hydrogel was prepared (Scheme 1). In this protocol, firstly the MNPs were prepared in the presence of montmorillonite clay (mMT). The magnetic nanoclay was then modified by CMC-g-Am through in situ graft copolymerization of Acrylamide by using KPS as a free radical initiator, and MBA as a crosslinking agent.
FTIR and XRD analysis
The FT-IR spectroscopy was carried out to confirm the chemical structures of synthesized hydrogels. Fig. 1a represents the spectra of CMC, CMC-g-Am, MT and mMT samples respectively. The FT-IR spectrum of pure CMC presents the broad absorption band at 3445cm-1 indicating the stretching frequency of the –OH groups. The bands at 2932 and 1065cm-1 represent the stretching vibration of C–H and C–O bonds, respectively. In the FT-IR spectrum of CMC-g-Am the presence of amide groups is confirmed by appeared bands at 3209 and 1611cm-1 correspondent to the stretching and bending of N–H in amide groups. According to the MT spectrum, the characteristic vibration bands of the clay appear at 3634cm-1 (–OH stretch of the lattice hydroxyl), 3448cm- 1 (–OH stretch from free H2O), 1643cm-1 (–OH bending) 1041cm-1 (Si–O stretch) and 522 and 460cm−1 (Si–O–Al and Si–O–Si bending vibrations) respectively. The FT-IR spectra of mMT demonstrated that by the introduction of MNPs into clay matrix due to van der Waals interactions between the oxygen groups of MT and MNPs, Fe–O bending vibrations of MNPswas shifted to low wave numbers. The bands between 436cm−1 and 615cm−1 were ascribed to Fe-O stretching vibration of MNPs which may be due to the overlapping of Si-O and Al-OH .
The crystalline structures and phase purities of as-synthesized hydrogels were studied by XRD patterns. The XRD patterns of pristine montmorillonite (MT) magnetic montmorillonite (mMT), CMC-g-Am, CMC-g-Am/mMT1 and CMC-g-Am/mMT2 hydrogels were shown in Fig. 1b. In the spectrum of neat MT a diffractive peak at 2θ=7.6° was observed which corresponds to the layered MT clay with d-spacing of 11.61 A0. Compared to the XRD pattern of MT, the characteristic peak at 2θ=7.6° was disappeared in the XRD pattern of mMT indicating the exfoliation of MT. Notably, during the synthesis of magnetic montmorillonite, the cation exchange can occurr between Na+ and Fe3+ (or Fe2+), and the synthesized MNPs can be located between clay sheets as well as on the surface of clay . The six characteristic peaks of MNPs appeared at 2θ about 30.4°, 35.4°, 43.5°, 53.3°, 57.4°, and 63.2° illustrating the corresponding indices (220), (311), (400), (422), (511), and (440), respectively. The interplaner distances were calculated according to Bragg,s equation and found to be 2.91Å (2θ=30.4°), 2.523 Å (2θ=35.4°), 2.086 Å (2θ=43.5°), 1.706 Å (2θ=53.3°), 1.606 Å (2θ=57.4°), and 1.47 Å (2θ=63.2°). The results are in accord with the database in JCPDS file (PDF No. 65-3107) indicating the formation of highly crystalline and pure magnetite nanoparticles with spinel structure. The broad peak at 2θ=7-17º may be attributed to the silicate composition of MT . The XRD patterns of CMC-g-Am, CMC-g-Am/mMT1 and CMC-g-Am/mMT2 hydrogels are depicted in Fig. 1b as well. The CMC-g-Am hydrogel displayed a common broad diffraction peak at (2θ=20°) illustrating the amorphous nature of PAM and the carbohydrate . The same diffraction peak with low intense was observed in nanocomposite hydrogel containing magnetic nanoclay, CMC-g-Am/mMT1 and CMC-g-Am/mMT2 suggesting that the clay layers are fully or partially exfoliated.
Field emission scanning electron microscopy (FE-SEM) coupled with energy dispersive spectroscopy (EDS) offers a semiquantitative elemental analysis of the surface of materials. At the first step, the surface morphology of dried hydrogels was exploited by FESEM technique and micrographs were shown in the Fig. 2a-c.
From SEM images, it was found that the quantity of (mMT) plays a drastic role in changing the morphology of the products. The free magnetic-clay hydrogel (CMC-g-Am) showed a relatively smooth and tight surface (Fig.2a). Interestingly incorporation of magnetic montmorillonite (mMT) significantly affected the morphology of hydrogels. As could be seen from Fig. 2b-c a rough structure was obtained for magnetic hydrogels, confirming the hydrogel intercalation in interlayer galleries of mMT. This observation is similar to our previous works using Na-MMT as nanoclay . Furthermore, the presence of magnetic nanoclay (mMT) in composition of hydrogel was confirmed by EDS spectra. EDS analysis on the other hand confirmed the formation of MNPs and MT by peak areas of (Fe and O) and (Na, Si, S) atoms in the hydrogel matrix. The EDS of CMC-g-Am, CMC-g-Am/mMT1 and CMC-g-Am/mMT2 was shown in Fig. 2a-c respectively. The characteristic peaks of Fe and Si appeared in the EDS spectra of both magnetic hydrogels. According to data, the content of Fe and Si in CMC-g-Am/mMT1 was obtained 2.38 and 12.73 %wt respectively. The content of Fe and Si in CMC-g-Am/mMT2 was 3.56 and 28.34 % wt respectively. This observation was in agreement with initial mMT used in synthesis process .
VSM and TEM studies
The hysteresis loops of CMC-g-Am/mMT hydrogel were measured by VSM technique between ±9 kOe at 298 K shown in Fig.3a. The magnetization curves illustrated that both CMC-g-Am/mMT1 and CMC-g-Am/mMT2 hydrogels were superparamagnetic with no coercively at room temperature. The saturation magnetization values for CMC-g-Am/mMT1 and CMC-g-Am/mMT2 hydrogels were 0.55 and 0.72 emu g-1, respectively indicating that magnetic intensity of the CMC-g-Am/mMT hydrogel can be tuned by altering the weight ratio of MNPs . Meanwhile, TEM image of CMC-g-Am/mMT2 depicted the presence of the nano-network architecture, shown in Fig.3b. The MNPs were found to be embedded with clay plates where no single nanoparticles were detected inside the hydrogel network.
Fig.4a represents the dynamic swelling behavior of magnetic hydrogels in distilled water. Obviously, the swelling degree of hydrogels exhibited a systematic trend in accordance with composition, in which the water uptake decreases with increased ratio of magnetic nanoclay. The swelling capacity of magnetic hydrogels was obtained lower than the free magnetic-clay hydrogels Maximum and minimum swelling capacity was obtained for free free-magnetic-clay CMC-g-Am and CMC-g-Am/MT2 hydrogels, respectively. This decrement in water absorbency can be attributed to the presence of MNPsand MT nanoclay as well. From one hand, the introduction of MNPs will cause (a) decrease in the ratio of hydrophilic and anionic functional groups and (b) the interaction between MNPs and polymeric chains can result in more crosslink points and thereby decrease in swelling capacity . On the other hand, by increasing the contents of MT the swelling ratios of hydrogels gradually decreased. Two phenomenons are involved here. Firstly, increasing the clay content can lead to an enhancement in ionic osmotic pressure of hydrogels arising from the mobile ions on the nanoclay particles . Secondly, nanoclay can act as a multifunctional crosslinker which subsequently limits the gel volume expansion . Our results are in accord with the previous studies. Cyras et alreported that the water absorption rate decreased after introduction of MT into hydrogel composition . Almasi et alalso observed an improved water resistance in starch/carboxymethyl cellulose composite films by addition of the nanoclay, ascribing to the stronger hydrogen bonds between carboxylic groups of CMC and OH of the MT layers . Yin et al have proposed that the swelling kinetics of cross-linked hydrophilic hydrogels are compatible with the Schott,s second-order diffusion models indicating that the swelling of hydrogels is controlled by stress relaxation of the network . Accordingly, the experimental swelling data were analyzed by non-linear second-order swelling kinetic according to Schott’s models described as Eq. 4 respectively .
Where, Wt is the water uptake at time t; We2 is the theoretical equilibrium water uptake for second-order kinetic, and k2 is the specific rate constant for the second-order swelling kinetic. The coefficient of determination (r2) was utilized to evaluate the fitting of the experimental data by the second-order swelling model . The model calculations were illustrated in Table 1. The value of r2 (≥0.98) corroborated the well-fitting of data by second-order kinetic. Additionally, the calculated equilibrated water absorbency (We2) for hydrogels was close to the experimental equilibrium water absorbency (We,meas) showing that the swelling rate is controlled by the stress relaxation in the network of hydrogel .