1. INTRODUCTION
Ulcerative colitis (UC) is a multifactorial intestinal disease caused by chronic inflammation. UC is characterized by severe intestinal inflammation and superficial mucosal ulcers [1–3]. The incidence of UC has more than doubled in several countries over the past decade [4]. UC has the clinical characteristics that include recurrent attacks, a low cure rate, a high risk of cancer, and a poor prognosis. US has been listed as one of the modern refractory diseases by the World Health Organization [5–7]. The etiology of UC is thought to be multifactorial, and includes a genetic predisposition, environmental factors, alterations in the immune system, disruption of intestinal epithelial integrity, and intestinal microecologic imbalances [8]. The current clinical treatments for UC include 5-aminosalicylic acid, salazosulfapyridine, glucocorticoids, antibiotics, and immunosuppressive agents, all of which have limitations [9, 10]. Therefore, the development of effective bioactive compounds is clearly needed for UC patients.
Ramulus mori (Sangzhi) alkaloids (SZ-A) are derived from a traditional Chinese medicine (Sangzhi), which is approved by the National Medical Products Administration of China for the treatment of type 2 diabetes mellitus (Approval no. Z20200002) [11]. SZ-A is a group of potent polyhydroxyl alkaloids, including 1-deoxyno-jirimycin (1-DNJ), fagomine (FAG), 1,4-dideoxy-1, 4-iminod-d-arabinitol (DAB), and other soluble polyhydroxyl alkaloids or glycosides with similar structures that have higher selective inhibition on intestinal glycosidase. Modern pharmacologic research has shown that when compared to single-target chemical drugs, SZ-A retains the advantages of multiple pharmacologic effects of natural drugs and has the characteristics of lowering body weight, regulating dyslipidemia, improving intestinal-pancreatic islet axis function, and lowering the body’s inflammatory state of the body [12, 13]. In addition, a large number of studies have shown that SZ-A acts on a variety of inflammatory pathways with anti-inflammatory activity [14]. Due to the short retention time of SZ-A in colonic tissues, the blood concentration is not maintained for a long period of time, which necessitates pre-administration modification or formulation design.
Natural polysaccharides have been widely used as carrier materials for drug delivery to lesion sites due to their high stability, biodegradability, and accessibility. Sodium alginate (SA) is a natural polysaccharide found in brown algae [15] that has been widely used in pharmaceuticals, food, and printing and dyeing because of its good pellet formation, biocompatibility, and bioadhesive properties [16–18]. SA applications in the biomedical field mainly include drug carriers, tissue engineering, and implantation of protective living cells [19, 20]. Chitosan (CS) is a kind of natural polysaccharide that is widely used in biomedical fields, such as tissue engineering and drug delivery systems [21, 22]. In addition, CS forms polyelectrolytes with alginate through electrostatic interactions, which not only compensates for the shortcomings of alginate hydrogels but also improves release properties and enhances bioadhesion properties [23]. In summary, SA-CS is a candidate material to prepare microspheres, encapsulate SZ-A, and modify SZ-A microspheres for the treatment of UC.
Most of the commonly used UC therapeutic drugs are oral formulations. Due to the special physiologic conditions of the gastrointestinal tract, traditional oral formulations are unable to effectively deliver drugs to the colon [24]. An in situ rectal thermosensitive hydrogel (P407-SA) benefits from the strong adsorption capacity of rectal mucosa and has a controllable drug release rate. P407-SA is liquid at room temperature and can be transformed into a semi-solid when it reaches the gelling temperature. P407-SA rapidly adheres to the lesion site to exert efficacy with a long retention time and achieves positioning targeting and continuous drug delivery [25–28]. The gel plug prepared by Ozguney et al. has good temperature sensitivity and bioadhesiveness [29]. Ramadass et al. prepared a mesalazine pH- and temperature-sensitive in situ rectal hydrogel for the treatment of UC [30].
In the current study the effects of drop injection, endogenous emulsification, and exogenous emulsification on the microstructure of SA-CS microspheres were investigated and the encapsulation rate and drug loading capacity of SZ-A microspheres were determined using SA-CS as the carrier material. In addition, temperature-sensitive hydrogels embedded with the SA-CS microspheres (TMH) system were constructed to provide a nnew direction for UC treatment.
2. MATERIALS AND METHODS
2.1 Materials
SA, CS, and nanoscale CaCO3 were purchased from Macklin (Shanghai, China). Poloxham 407 (P407) was purchased from BASF (Ludwigshafen, Germany). Pectin and CaCl2 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Glutaraldehyde was purchased from J&K Scientific (Beijing, China). TPP was purchased from Aladdin (Shanghai, China). NaHCO3 and C6H9Na3O9 were purchased from Sinopharm (Beijing, China). Dialysis bags were purchased from Solarbio (Beijing, China).
2.2 Preparation of SA-CS microspheres
2.2.1 Preparation of SA-CS microspheres using the droplet injection method
SA-CS microspheres were prepared using the drip method with CS and SA as raw materials. Briefly, SA (10 mg/mL) was slowly dripped into a mixture of CaCl2 (20 mg/mL) and 1% CS (w/v) acetic acid, magnetically stirred for 1 h, filtered, washed, and dried to obtain non-crosslinked SA-CS microspheres. After cross-linking of genipin (10 mg/mL) and CS in a 37°C water bath for 24 h, when compared to the preparation of non-crosslinked SA-CS microspheres, the other steps were performed as above to obtain the crosslinked SA-CS microspheres of genipin. In preparing composite microspheres loaded with SZ-A, when compared to the preparation of blank microspheres, it is necessary to add SZ-A in different mass ratios (5:1, 2:1) to the prepared SA solution with sufficient stirring. The rest of the preparation remains unchanged, which leads to the composite microspheres loaded with SZ-A.
2.2.2 Preparation of SA-CS microspheres using the internal emulsification method
SA-CS microspheres were prepared by internal emulsification with CS and SA. The SA solution (10 mg/mL) was preheated to 40°C and 0.08 g of nanoscale CaCO3 and SZ-A were added, stirred thoroughly, and mixed well as the aqueous phase. Liquid paraffin containing 1% Span80 and 0.5 mL of glacial acetic acid was used as the oil phase and stirred for 30 min at 40°C under water bath conditions. The aqueous phase was added dropwise to the above oil phase solution at 40°C, mechanically stirred for 30 min, emulsified for 15 min, and acid-initiated for 15 min to form SA microspheres. CS solution (4 mg/mL) was added to the above-mixed solution and the membrane was overlaid for 30 min. When the reaction was complete, the layers were left to stratify and the lower precipitate was removed, centrifuged, and washed with petroleum ether, anhydrous ethanol, and purified water three times each to form composite microspheres.
2.2.3 Preparation of SA-CS microspheres using the exogenous emulsification method
SA-CS microspheres were prepared from CS and SA using the extrinsic emulsification method. SA solution (10 mg/mL) was fully mixed with SZ-A as the water phase and liquid paraffin containing different concentrations of Span80 was used as the oil phase. SA solution was slowly added to the oil phase solution at 40°C with stirring and emulsifying for 30 min to form a water/oil (W/O) emulsion. The CS and CaCl2 mixture was slowly added to the emulsified mixture and cross-linked for 2 h. Two percent glutaraldehyde (GA) solution was slowly added to the above-mixed solution and cross-linked for 1 h to cure. After the reaction was complete, the lower precipitate was removed, centrifuged at 6000 rpm for 10 min, and washed 3 times with petroleum ether, anhydrous ethanol, and purified water to form composite microspheres, which were then dried and stored.
2.3 Characterization of SA-CS microspheres
SA-CS microspheres were freeze-dried and the images were obtained in the 15 KV-secondary electron mode. The surface morphology of the dispersion was observed under a scanning electron microscope [SEM] (Thermoscientific ApreoS, FEI, Waltham, MA, USA). The structure and composition of SA-CS microspheres were analyzed by Fourier transform infrared spectroscopy [FT-IR] (Nicolet, Madison, WI, USA), differential scanning calorimetry [DSC] (Netzsch, Bavaria, Germany), and X-ray diffraction [XRD] (Bruker D8 Advance, Karlsruhe, Germany).
2.4 Construction of TMH system
Using P407 as the carrier material, the P407 concentration was maintained at 15% (w/w), different amounts of SA (0.3%–1.2%) were added, and the TMH system was refrigerated at 4°C for 24 h to fully swell and obtain a clarified and evenly dispersed P407-SA hydrogel solution.
2.5 Determination of the gelling temperature (Tgel)
The Tgel of the thermosensitive hydrogel was determined by the tilt method. Hydrogel (2 mL) was measured in a test tube, a precision thermometer was inserted, and a constant temperature water bath was opened. The temperature in the water bath was maintained for 5 min and when the temperature increased by 0.1°C, the inverted test tube was tilted for a 15 s-interval. The Tgel is the reading on the thermometer when the hydrogel presents a semi-solid state.
2.6 Rheologic characterization
Unlike ordinary hydrogels, temperature-sensitive hydrogels have viscous liquid and elastic solid rheologic properties. The viscous liquid properties are expressed in terms of the rheologic parameter (viscous modulus G″). The elastic solid properties are expressed in terms of the rheologic parameter (elastic modulus G′). When G′ > G″, the temperature-sensitive hydrogel exhibits an elastic-based solid-like state. When G′ < G″, the temperature-sensitive hydrogel exhibits viscous-based fluidic properties. The viscoelastic interval of the hydrogel was determined by amplitude scanning (strain, 0.1%–100%; frequency, 1 Hz) using a Kinexus Lab+ rotary rheometer (Netzsch). The hydrogel modulus change was measured by frequency scanning (strain, 0.5%; frequency, 0.1–10 Hz). In addition, the pre- and post-hydrogel formation viscosities of the hydrogels were determined.
2.7 In vitro drug release
The drug release behavior of SZ-A, SA-CS microspheres, and TMH was evaluated in a simulated colonic fluid. SA-CS microspheres (10 mg) were dissolved in 2 mL of ultra-pure water. The release behavior of SZ-A in TMH followed the addition of SA-CS microspheres (10 mg) to 2 mL of temperature-sensitive hydrogel solution. The TMH was solidified into a semi-solid by placement in a constant temperature water bath at 37°C before release. The drug release process of SA-CS microspheres and TMH in vitro was studied using the dialysis bag method. SA-CS microspheres and TMH were placed into a dialysis bag, which was in turn placed in a centrifuge tube with 20 mL of simulated fluids at 37°C and centrifuged at 100 rpm. Samples (2 mL) were withdrawn periodically with the same volume of fresh dissolution medium and the amount of drug was determined by high-performance liquid chromatography.
3. RESULTS AND DISCUSSION
3.1 Preparation of SA-CS microspheres using the droplet injection method
SA is a polyanionic natural polysaccharide with an abundant reserve, a low price, and excellent biocompatibility, degradability, and sphericity. SA is widely used in pharmaceutical, food, textile, and other industries. The Na2+ on the SA is easily exchanged with divalent metal cations (Ca2+, Ba2+, and Sr2+), which forms a three-dimensional network of molecules in the “eggshell” structure of the gel. CS is a deacetylated derivative of chitin with a cationic charge and has good biocompatibility, degradability, and adhesion properties. The negatively charged carboxyl group in SA and the positively charged amino group in CS form polyelectrolyte complexes due to electrostatic interactions. In the current study SA-CS hydrogel microspheres were prepared using SA-CS as the carrier material to determine the effects of different preparation methods on the morphology, particle size, encapsulation rate, and drug loading capacity of hydrogel microspheres. The drip preparation method is the earliest and most widely used. First, SA-CS microspheres were prepared using the drop injection method and the effects of CS concentration, cross-linking agent, different carrier materials, and feeding ratios on the morphology and encapsulation rate of SA-CS microspheres and drug loading capacity were investigated.
The droplet method involves the preparation of microspheres by dropping SA into a cation-containing solution, the particle size of which is affected by the aperture of the needle. First, the effects of different needle apertures on the particle size of the microspheres were investigated ( Figure 1 ). The experimental results showed that the microsphere particle size decreased with a decrease in the syringe diameter. When the syringe pore size was too small (32 GB), the microspheres did not drip easily. Therefore, the pore size of a 1-mL syringe was selected for the study.
Different concentrations of CS, the feed ratio, and the crosslinker affect the encapsulation rate and microsphere drug loading. The feeding ratio of SA to SZ-A was fixed at 5:1 and the effects of different concentrations of CS (0.2%, 0.4%) on the microsphere encapsulation rate and drug loading were investigated. Genipin acts as a cross-linking agent that reacts with CS in a crosslinking reaction, resulting in a polymer that emits green fluorescence under excitation of a multi-band light source. SA-CS microsphere cross-linking with genipin was observed under a fluorescence inverted microscope; the experimental results are shown in Figure 2 . It can be seen in Figure 2 that the surface of the microsphere is a green ring, indicating that genipin was successfully crosslinked to the surface of the CS. It can be seen from Figure 2 that the particle size of the microspheres exceeded 1000 μm because the SA-CS microspheres prepared using the drop injection method were affected by the aperture of the needle, which makes the particle size too large. The effect of the addition or absence of genipin on the microsphere encapsulation rate and drug loading was also investigated in this experiment; the experimental results are shown in Table 1 . The encapsulation rate and microsphere drug loading increased with an increase in the CS concentration when genipin was not used because an increase in the CS concentration thickened the shell formed on the surface of the microspheres, which reduced the loss of drug when the microspheres were washed in water, resulting in an increase in the encapsulation rate and drug loading capacity of the microspheres. However, the microspheres did not have a high encapsulation rate or drug loading capacity because SA has water absorption and easy swelling characteristics, which result in a larger pore size when SA-CS microspheres are washed and SZ-A is released from the inside to the outside, resulting in a low encapsulation rate and microsphere drug loading. Addition of the crosslinking agent, genipin, did not improve the encapsulation rate or microsphere drug loading. Therefore, subsequent experiments did not incorporate genipin.
Effects of CS on SZ-A loading and encapsulation rate of SA-CS microspheres.
| SA:SZ-A | CS (%) | Crosslinking | LE (%) | EE (%) |
|---|---|---|---|---|
| 5:1 | 0.2 | √ | 0.68 ± 0.058 | 3.66 ± 0.31 |
| 5:1 | 0.4 | √ | 0.61 ± 0.04 | 3.32 ± 0.22 |
| 5:1 | 0.2 | / | 0.77 | 4.40 ± 0.04 |
| 5:1 | 0.4 | / | 1.16 ± 0.02 | 10.98 ± 0.16 |
Pectin, a natural polymeric polysaccharide polymer, was added to enhance the drug loading capacity of SA-CS microspheres. The ratio of SA to SZ-A was increased to 2:1, and the effects of SA, CS, and pectin on microsphere drug loading were investigated. The experimental results are shown in Table 2 . The drug loading and encapsulation rate of the microspheres were increased when SA, CS, and pectin were used in combination. Moreover, the concentration of pectin was increased; the experimental results are shown in Table 3 . The results showed that increasing the concentration of pectin did not significantly improve the microsphere drug loading. In general, the drug loading and encapsulation rate of SA-CS microspheres prepared using drip method were not high and there was no significant improvement, even with an increase in the pectin concentration. This finidng may be due to the large leakage of SZ-A into the aqueous environment during the preparation process with less drug encapsulated in the microspheres.
3.2 Preparation of SA-CS microspheres using the internal emulsification method
Due to the low encapsulation rate and drug loading capacity of microspheres obtained using the injection-drop method, the preparation of SA-CS microspheres by endogenous emulsification was investigated. The endogenous emulsification method involves dispersing CaCO3 powder in an SA aqueous solution, then emulsification to form SA droplets containing the powder. The acidic aqueous solution was dripped into the oil phase, where H- gradually diffused into the interior of the droplets and reacted with CaCO3 to produce Ca2+, which gradually diffused outwards to solidify the SA droplets and form SA-CS microspheres. The results of the endogenous emulsification method are shown in Table 4 . The results showed that the microsphere encapsulation rate and drug loading capacity obtained from the microsphere prepared by the endogenous emulsification method were increased compared to the droplet method. Overall, the encapsulation rate of SA-CS microspheres was still low, which may be due to the more porous structure of the microspheres prepared by endogenous emulsification.
3.3 Preparation of SA-CS microspheres using the exogenous emulsification method
Unlike endogenous emulsification, the exogenous emulsification method forms a W/O emulsion from the SA aqueous solution and oil phase, then the aqueous solution containing Ca2+ is dripped into the emulsion and Ca2+ gradually diffuses from the outside of the SA droplet into the inside of the droplet, curing from the surface to the inside. This method is produced on a larger scale and can produce microspheres with smaller particle sizes.
3.3.1 Effect of the emulsifier concentration
The concentration of fixed CaCl2 was 5%, the concentration of CS was 0.2%, and the ratio of water-to-oil was 1:2. The effects of Span80 dosage (1%–5%) on the encapsulation rate and drug loading of SA-CS microspheres were investigated. The experimental results are shown in Table 5 . When the production of Span80 was 1% or 4%, the drug loading and encapsulation rate of the prepared microspheres were low. The encapsulation rate and microsphere drug loading increased when the ratio of Span80 in the oil phase is 2% or 5% (v/v). Therefore, a Span80 ratio of 2% or 5% was selected for the subsequent optimization of the prescription of microspheres.
3.3.2 Effect of the oil-to-water volume ratio
The concentration of fixed CaCl2 was 5%, the concentration of CS was 0.2%, and the ratio of Span80 in the oil phase is 2% or 5% (v/v). The effects of the volume ratio of the water-to-oil phase (1:1, 1:2, and 1:4) on the encapsulation rate and drug loading of SA-CS microspheres were investigated. The experimental results are shown in Tables 6 and 7 . The mixed solution was not sufficiently emulsified to form a stable emulsion when the oil-to-water ratio was 1:1. When the oil-to-water ratio was 2:1, the encapsulation rate and drug loading of the prepared microspheres were higher. The drug loading and encapsulation rate of the prepared microspheres decreased when the oil-to-water ratio was 4:1.
3.3.3 Characterization of the SA-CS microsphere morphology
The effect of the Span80 ratio (1%–5%) on the morphology of SA-CS microspheres was investigated when the concentration of CaCl2 was 5%, the concentration of CS was 0.2%, and the ratio of the water-to-oil phase was 1:2. The experimental results are shown in Figure 3 . When the Span80 ratio was 1% or 4%, the surface of the prepared microspheres was rough and the sphericity was poor. The microsphere formation rate of the prepared microspheres was relatively high when the Span80 ratio was 2% or 5%. Overall, the microsphere surface was uneven and poorly spherical.
SEM images of SA-CS microspheres prepared with different ratios of Span80; the volume ratio of the water-to-oil phase was 1:2.
The morphology of SA-CS microspheres prepared at different rotational speeds was determined. SA-CS microspheres were prepared by fixing the above preparation conditions when the volume ratio of the water-to-oil phase was 1:2, the ratio of Span80 in the oil phase is 2% (v/v), and the rotational speeds were changed to 400, 600, and 750 rpm. The experimental results are shown in Figure 4 . The results showed that when the rotational speed was increased from 400 to 750 rpm, the prepared microspheres all had a specific rate of ball formation. When the rotational speed was too large, the probability of collision and agglomeration of microspheres during the curing process was increased, resulting in instability of the emulsion and difficulty in forming monodisperse droplets. When the rotational speed was too low, SA-CS microspheres cannot be formed. Therefore, 600 rpm was selected as the preparation condition for SA-CS microspheres.
SEM images of SA-CS microspheres prepared under different stirring speeds; the ratio of Span80 was 2% and the volume ratio of the water-to-oil phase was 1:2.
The fixed concentration of CaCl2 was 5%, the concentration of CS was 0.2%, and the ratio of the water-to-oil phase was 1:4. The effects of the Span80 ratio (2%, 5%) on the morphology of SA-CS microspheres were investigated. The experimental results are shown in Figure 5 . The results showed that when the ratio of water-to-oil was 1:4 and the ratio of Span80 in the oil phase is 5% (v/v), the prepared microspheres were round with a high pellet formation rate, drug loading of 19.80%±0.028, and encapsulation rate of 30.19%±0.075. Therefore, the water-to-oil ratio of 1:4 and the ratio of Span80 of 5% were selected as the optimal conditions for the preparation of SA-CS microspheres.
3.3.4 Effect of GA concentration
To reduce the pores on the surface of SA-CS microspheres and increase densification of the microsphere surface, the effect of different concentrations of GA crosslinking on microsphere drug loading rate was investigated. The experimental results are shown in Table 8 . The results showed that the microspheres were prepared with a higher drug loading (17.56% ± 0.259) when 2% GA (1 mL) was added. Further, the effects of GA crosslinking at different concentrations on the morphology of microspheres were investigated; the experimental results are shown in Figure 6 . The results showed that the surface pores of the microspheres decreased and the smoothness increased as the crosslinker concentration increased. Combining these results, 2% GA (1 mL) was selected as the optimal condition for SA-CS microsphere preparation.
3.3.5 Characterization of SA-CS microspheres
FT-IR spectroscopy was used to evaluate the structure of SA-CS microspheres; the obtained spectra are shown in Figure 7A . As shown in Figure 7A , the strong and broad peak of SA at 3219 cm−1 is the telescopic vibrational absorption peak of the -OH hydrogen bonding linker. The anti-symmetric and symmetric telescopic vibrational absorption peaks of -COO- were 1593 and 1404 cm−1, respectively. The stretching vibration absorption peak of -OH was 1023cm−1. The -OH absorption peak of CS was 3355 cm−1, the CH3 telescopic vibrational absorption peak was 2865 cm−1, the characteristic bending vibrational absorption peak of amino-NH2 was 1587 cm−1, and the CH3 bending vibrational absorption peak was 1373 cm−1. When SA undergoes ionic gelation with Ca2+, the -OH stretching vibrational absorption peak on SA is red-shifted and appears at a lower wave number, which may be due to the interaction of part of Ca2+ with -OH on SA. The bending vibration of the six-membered ring -OH is restricted in SA-CS microspheres compared to SA, as evidenced by the disappearance of the absorption peak at 945 cm−1 and weakening of the absorption peak at 883 cm−1 in the direction of the low wave number. After the SA-CS microsphere was loaded into the SZ-A, the characteristic peak of the SZ-A disappeared, indicating that the SZ-A had been encapsulated in the microsphere.
Characterization of SA-CS microspheres.
(A): Infrared spectrum of SA-CS microspheres; (B): DSC spectra of SA-CS microspheres; (C): XRD pattern of SA-CS microspheres. A: a SA; b CS; c SZ-A; d SA-CS microspheres; e SA-CS-SZ-A microspheres; f GA-SA-CS-SZ-A microspheres. B: a SA-CS physical mixture; b SZ-A; c SA-CS microspheres; d SA-CS-SZ-A microspheres; e GA-SA-CS-SZ-A microspheres. C: a SA; b CS; c SA-CS microspheres; d SA-CS-SZ-A microspheres; e GA-SA-CS-SZ-A microspheres.
Figure 7B shows the DSC pattern of SA-CS microspheres. The SA thermal decomposition process is divided into the following two stages: SA loses the intramolecularly bound water; and there will be a broad heat-absorption peak near 100.53°C. The SA molecular chain breaks to produce a stable intermediate product that will show a sharp exothermic peak near 246.41°C. The peak at 311.40°C is the exothermic peak of CS, which is due to the degradation of the polyelectrolyte. Compared to the SA-CS blends, the exothermic peaks of the SA-CS blank microspheres had a tendency to flatten out, whereas the microsphere plot did not show the exothermic peak of CS at 311.40°C. The reason for these changes may be the result of the joint action of SA and CaCl2. Compared to the SA-CS-SZ-A microspheres, the heat absorption peak of the crosslinked microspheres was pushed back to 163.27°C, while a sharp heat absorption peak appeared at 211.79°C, which might be due to crosslinking of CS with GA. The characteristic absorption peak of SZ-A disappeared in the drug-loaded microspheres, indicating that SZ-A had been encapsulated in the microspheres.
Figure 7C shows the XRD pattern of SA-CS microspheres. It can be seen in Figure 7C that CS has strong intermolecular hydrogen bonding and is strongly crystalline, with sharp crystalline peaks at approximately 10.4° and 20°. Because SA is a block polymer, SA does not have obvious diffraction peaks like crystalline materials and the characteristic peaks are at approximately 13° and 21°. As shown in Figure 7C , crosslinking of CaCl2 has a destructive role in the crystalline structure of SA, attenuating the diffraction peaks of SA, which approaches an amorphous structure that may be due to the formation of a network eggshell structure of SA after crosslinking of CaCl2. The XRD of SA-CS microspheres is relatively flat, with broadening at 11° and 20°, which may be the result of the crosslinking reaction and the strong electrostatic gravitational force between SA and CS. This finding indicates that there is a strong electrostatic interaction between CS and SA that destroys the arrangement order of CS molecules. SA-CS has an amorphous state and the crosslinked microspheres lose the original crystalline structure.
3.4 Construction of the TMH system
Due to the special physiologic environment of the gastrointestinal tract, traditional oral formulations do not effectively deliver drugs to the colon and can produce adverse reactions with weakening of the therapeutic effect on colonic diseases. As an effective therapeutic strategy for intestinal diseases, rectal administration can reduce systemic side effects by reducing systemic absorption of the drug and maintaining a high drug concentration in the colon to improve the local bioavailability of the drug. Medication compliance is also improved. In this experiment the TMH system was established and the suitable prescription for rectal administration was screened with the Tgel as the index (32 ± 0.5°C). The mass ratio of P407 in the total system is 15% (w/w) and the mass ratio of SA in the total system is 0.3%-1.2% (w/w) added to prepare the TMH. The Tgel of TMH was determined using the test-tube inversion method. The measurement results of the Tgel are shown in Figure 8 . The Tgel decreased with an increase in the SA ratio. This finding may reflect the enhanced hydrogen bonding interactions between SA and P407 with an increase in the mass ratio of SA in the total system is 0.8% (w/w). The TMH was more likely to form hydrogels at lower temperatures. Taking the ideal Tgel as the index, the appropriate prescription ratio of rectal administration was as follows: SA, 0.8%; and P407, 15%.
3.5 Rheologic study
The maximum deformation of the internal three-dimensional network structure of the hydrogel was obtained by the results of the hydrogel linear viscoelastic region (LVR) test. Amplitude scanning can be used to characterize the stability of hydrogel structures. If G′ does not change, the hydrogel is not destroyed and all determinations need to be made within the linear viscoelastic zone. The measurement results of the hydrogel LVR are shown in Figure 9A . The hydrogel had a wide linear viscoelastic zone and a more stable internal structure in the 0.1%–100% strain scanning range. The result of the frequency scan is shown in Figure 9B . The G′ of the hydrogel remained stable in the frequency scanning range of 0.1–10 Hz and did not show a significant frequency dependence. Furthermore, the viscosity of the hydrogel before and after the phase transition was determined. The experimental results are shown in Figure 9C and D . The hydrogel had a low viscosity at room temperature and an increased viscosity at body temperature, achieving a phase transition from liquid-to-solid.
Rheologic characterization of thermosensitive hydrogels.
(A): Linear viscoelastic region scanning of thermosensitive hydrogels; (B): Frequency scanning of thermosensitive hydrogels; (C): Viscosity curve of temperature-sensitive hydrogel with shear rate at room temperature; (D): Viscosity curve of thermosensitive hydrogel with the shear rate at the gelling temperature.
3.6 In vitro release study
The release behavior of SZ-A, SZ-A-loaded microspheres, and thermosensitive hydrogels is shown in Figure 10 . A faster release rate was observed for the microspheres at the first two time points, which could be attributed to the rapid dissolution of the drug adhering to the surface of the microspheres. In contrast, the microspheres can be released rapidly in the colonic fluid environment, which is attributed to the pH sensitivity of SA. It has been reported that the swelling and degradation rates of polymers directly affect drug release kinetics. However, polymer degradation is not the only mechanism that affects drug release. Drug release is also controlled by diffusion because the drug diffuses through the pores of the microspheres. Porosity has a significant effect on drug release; the greater the porosity, the faster the drug release. The drug release rate of the microspheres encapsulated in the thermosensitive gel was clearly weakened because the state of the thermosensitive gel rapidly changed from a flowing liquid state to a semi-solid state at the gelling temperature and the viscosity changed greatly, making the flow of the drug suspension slow down.
4. CONCLUSION
In the current study, using SA-CS as the carrier material, the effects of the drip, internal emulsification, and external emulsification methods on the microstructure of SA-CS microspheres were investigated, and the encapsulation rate and drug loading of the SA-CS microspheres were measured. On this basis, the TMH system was constructed for the treatment of UC. The results showed that the SA-CS microspheres prepared by the exogenous emulsification method had the highest encapsulation rate and drug loading capacity, better sphericity, and more uniform particle size. The optimum conditions for preparing SA-CS microspheres were determined by observing the morphology of the microspheres and measuring the drug loading, as follows: stirring speed, 600 rpm; ratio of water-to-oil phase, 1:4; and ratio of emulsifier Span80 in the oil phase is 5% (v/v). Characterization by FT-IR, DSC, and XRD revealed that SZ-A and SA were bound to each other in a physically encapsulated manner and SZ-A was dispersed in SA-CS microspheres as amorphous crystals. Furthermore, the TMH system was constructed to enable rapid adhesion of the drug to the colon. The thermosensitive hydrogel was liquid at room temperature and rapidly transformed into a semi-solid when the gelling temperature was reached, which can be attached to the lesion site. Rheologic results indicate that the thermosensitive hydrogels have a stable internal structure with shear-thinning properties. The results of in vitro release showed that the thermosensitive hydrogel-containing microspheres had a sustained release effect and prolonged the retention time of the drug in the colon. In conclusion, these findings provide ideas for the preparation of SA-CS microspheres and the design of other thermosensitive hydrogel formulations.
