Ionic liquid-assisted cellulose coating of chitosan hydrogel beads and their application as drug carriers

The present study proposes a simple yet effective method of cellulose coating onto chitosan (CS) hydrogel beads and application thereof as drug carriers. The beads were coated with cellulose dissolved in 1-ethyl-3-methylimidazolium acetate, an ionic liquid (IL) via a one-pot one-step process. Water molecules present in the CS beads diffused outward upon contact with the cellulose–IL mixture and acted as an anti-solvent. This allowed the surface of the beads to be coated with the regenerated cellulose. The regenerated cellulose was characterized by FE-SEM, FT-IR, and XRD analyses. To test potential application of the cellulose-coated CS hydrogel beads as a drug carrier, verapamil hydrochloride (VRP), used as a model drug, was impregnated into the beads. When the VRP-impregnated beads were immersed in the simulated gastric fluid (pH 1.2), the VRP was released in an almost ideal linear pattern. This easily fabricated cellulose-coated CS beads showed the possibility for application as carriers for drug release control.

www.nature.com/scientificreports/ In this study, CS beads were coated with cellulose dissolved in IL using only water molecules contained in hydrogels ( Supplementary Fig. S1). The prepared cellulose coated CS beads were characterized by Field emission scanning electron microscope (FE-SEM), Fourier transform infrared spectrometer (FT-IR), and X-ray diffractometer (XRD), and applied to drug release experiments and discussed accordingly.

Results and discussion
Morphology of cellulose-coated cS bead. The conceptual strategy for the cellulose coating on hydrogel is shown in Fig. 1. The strategy was to evenly coat wet CS hydrogel bead with dissolved cellulose via osmotic pressure difference. That is, cellulose was first dissolved in [Emim][Ac], which is a well-known cellulose processing IL 29 . Then the wet CS bead, which was used as a model hydrogel, was brought into contact with the dissolved cellulose. Upon contact with the cellulose solution, the water molecules present inside of the CS bead diffused outwards due to osmotic pressure. Consequently, the diffused water molecules acted as anti-solvents to solidify the cellulose and formed cellulose coating on the surface of the hydrogel. It must be noted that control of the water content of the hydrogel bead was essential at this stage. It was observed that when the bead was too watery (having water on the outside), the cellulose coating formed was uneven, rough and tend to clump up. Moreover, when the bead was dried or had no water inside, it was impossible to form the cellulose coating due to the lack of water molecules to act as anti-solvents. Thus, the best condition for this idea to be realized was to control the moisture content of the bead by effectively wiping off surface water. For this purpose, the wet bead was placed on gauze to remove adhering water from the surface of the bead before use.
fe-SeM analysis. To understand the texture, the cellulose-coated CS bead was cut in half, and the morphology was analyzed using FE-SEM, as shown in Fig. 2. The FE-SEM images showed the apparent boundary of CS and cellulose layers (Fig. 2a). The cross-sectional internal morphology (Fig. 2b) clearly showed that regenerated cellulose appeared as an outer layer (shell), and CS hydrogel was present as the core. Higher magnification of core and shell parts showed a connected three-dimensional network ( Supplementary Fig. S2). Both the core and shell parts possessed high porosities, with the shell part showing a broad range of pore sizes and the core part revealing smaller pore sizes. This porous network can be useful to encapsulate drugs or cells for biological applications. In addition, the results of FE-SEM analysis (Fig. 3A) and element mapping ( Fig. 3B-D) of the crosssection of the cellulose-coated CS bead are shown in Fig. 3. Since carbon and oxygen are constituent elements of chitosan (CS) and cellulose, the elements C (Fig. 3B) and O (Fig. 3C) were evenly distributed throughout the cellulose-coated CS bead. However, it was confirmed that the elements of N (Fig. 3D) are concentrated in the core (chitosan part). Additionally, as a result of EDS mapping by separating the core part and the shell part, C, O,  www.nature.com/scientificreports/ N, and Na were measured as 70.32, 25.71, 3.88, 0.09 wt% and 53.64, 46.28, 0.00, 0.08 wt%, respectively in the core and shell parts (Table 1). That is, the N element was detected only in the core, the chitosan portion. Therefore, it was confirmed that the core is composed of chitosan and the shell is composed of cellulose. In addition, if some of the dissociated [Emim][Ac] remains in cellulose-coated CS bead, N, a constituent element of [Emim][Ac], must be detected not only in the core but also in the shell part. However, since the cellulose-coated CS beads were thoroughly washed, no ionic liquid residues were considered.
ft-iR analysis. Prior to FT-IR analysis, the cellulose-coated CS bead was separated into the shell (regenerated cellulose) and core (CS) part to confirm the functional groups of each section. FT-IR spectroscopic analysis of the original cellulose and regenerated cellulose on CS bead are shown in Fig. 4. The broad absorption band in the range of 3,300-3,400 cm −1 was due to the intermolecular O-H stretching vibration of the cellulose molecule. The peak at 2,898 cm −1 observed for native cellulose was due to the C-H stretching vibration of CH 2 and CH 3 groups. This band was not affected by changes of crystallinity; thus, there was no significant change of this band in the regenerated cellulose spectral data. The peak at 1,428 cm −1 in the original cellulose was assigned to the crystallized cellulose I and amorphous cellulose, whereas in the case of the regenerated cellulose, this band shifted to 1,419 cm −1 , representing cellulose II and amorphous cellulose 29,30 . The peak at 1,428 cm −1 in the cellulose spectra was assigned to symmetric CH 2 bending vibration. The peak at 1,103 cm −1 , which appeared in the spectrum of the native cellulose was not observed for the regenerated cellulose. This further confirmed the prevalence of crystalline cellulose II 7 . In the regenerated cellulose spectra, the new peak at 1,750 cm −1 was attributed to the C=O in the acetate 31 . This is good evidence that cellulose dissolved in   In vitro release of VPR from CS beads and cellulose-coated CS beads. Since cellulose coatings may improve the physical and chemical stability of the hydrogels, cellulose-coated CS beads may have a wide range of applications. One of the attractive applications of the cellulose-coated hydrogels could be in the biomedical area because both cellulose and CS are biocompatible and biodegradable. Hence we tested the cellulose-coated CS beads for in vitro verapamil hydrochloride (VRP) release applications. VRP is a calcium channel blocker used for the treatment of hypertension, angina and myocardial infarction 34 . The release experiments were studied in different release fluids, including simulated gastric fluid (SGF, pH 1.2) and simulated intestinal fluid (SIF, pH 6.8). The release profiles corresponding to a formulation consisting of CS beads and cellulose-coated CS beads are displayed in Figs. 6 and 7 for SGF and SIF, respectively. As shown in Fig. 6, the release patterns of VRP from the cellulose-coated CS beads in SGF were significantly  www.nature.com/scientificreports/ different from those of CS beads. This could be explained by the stability of the prepared CS hydrogels. In SGF, CS beads were dissolved only within 5 min, and no solid matter was left after 10 min in the system. Meanwhile, the cellulose-coated CS beads were stable and remained stable up to 24 h. The release of VRP from the cellulosecoated CS beads was almost linear, similar to the zero-order release. Figure 7 shows the VRP release profiles of the CS beads and cellulose-coated CS beads in SIF. The release of VRP from the CS beads was much faster than that from the cellulose-coated CS beads. After 120 min, 95 ± 0.5% and 65 ± 0.2% of VRP were released from the CS beads and cellulose-coated CS beads, respectively, indicating that the cellulose coating makes a significant change in VRP release profiles. The total amounts of drug released from both coated and uncoated CS beads after 300 min were nearly the same. The results of this study were tried to be applied to the well-explained drug release model Korsemeyer-Peppas model (supplementary information). The release exponent (n) for CS beads are smaller than 0.45 in SGF (0.0752) and SIF (0.3217), indicating that the release of VRP from the CS beads are a Quasi-Fickian diffusion mechanism (Supplementary Table S1). This mechanism indicated that VRP diffuses partially through a swollen matrix or its pores in the chitosan hydrogels. Moreover, the n values of cellulose-coated CS beads in SGF and SIF were 0.7298 and 0.6628, respectively, which indicated the non-Fickian diffusion mechanism consisting of a combination of diffusion and polymer relaxation. By coating cellulose on CS beads, it was possible to alter the release pattern of the drugs in the CS beads. These  www.nature.com/scientificreports/ results clearly demonstrated the potential application of the prepared cellulose-coated CS hydrogel beads in the biomedical area of drug delivery.

conclusions
Cellulose-coated CS beads were successfully fabricated via a facile one-pot one-step method. FE-SEM observation showed that the CS beads and the cellulose coating layer were separated and porous. The results of FT-IR and XRD also supported that the regenerated cellulose was well coated on the CS surface. The cellulose-coated CS beads exhibited sustained release patterns of VRP in SGF and SIF environments when applied as drug carriers. This simple cellulose coating method will be able to promote various applications of the hydrogels.

Fabrication of VRP-loaded CS beads.
To fabricate the VRP-loaded CS beads, 1,000 mg of VRP was added to 10 ml of 2% CS solution and stirred for 30 min to obtain a homogeneous dispersion. The VRP-loaded CS beads were prepared in the same dropwise method, as described in the previous section.
procedure for cellulose coating on cS beads. The wet CS beads were dropped into the 6% (w/v) cellulose dissolved in [Emim][Ac] solution and kept for 5 min, resulting in the formation of cellulose coat on the surface of CS beads. Then, the cellulose-coated CS beads were separated from the solution using gauze mesh and washed with deionized water three or more times.

Fabrication of VRP-loaded cellulose-coated CS beads. The VRP-loaded cellulose-coated CS beads
were fabricated by first manufacturing the VRP-loaded CS beads and then coating it with cellulose. The verapamil encapsulation efficiency (EE) and loading capacity (LC) of cellulose-coated CS beads used in drug release experiments were 100 ± 0.0% and 8.62 ± 0.07%, respectively. Encapsulation efficiency (EE) and loading capacity (LC) were calculated as follows: Drug release studies. The in vitro release studies were performed in simulated gastric fluid (SGF, pH 1.2) and simulated intestinal fluid (SIF, pH 6.8). Ten beads were placed in plastic bottles containing 50 ml of the release medium. After the cellulose coating, the weight per bead was increased by cellulose coating, so the number of beads was used instead of the weight of beads. The drug release experiments were carried out at 100 rpm and 37 ± 0.5 °C in a shaker. At predetermined time intervals, samples of 0.5 ml were collected from the release medium and replaced with fresh SGF and SIF solution, respectively. The experiments were duplicated. To estimate the initial concentration of VRP entrapped into the beads, the VRP concentration was measured after crushing and dissolving 10 VRP-loaded CS beads in 50 ml of medium, which is the same condition as the release experiment. instrumental details. In the dried cellulose-coated CS bead, the core and the shell part are easily separated. In this study, a cross-section was cut using a stationery knife and tweezers and then FE-SEM analysis was EE (%) = weight of loaded drug weight of total added drug × 100 LC (%) = weight of loaded drug weight of drug loaded bead × 100.