2.1. Synthesis and Characterization of β-CDP
As shown in
Scheme 1a, β-CDP was prepared by the reaction between the hydroxyl hydrogen of β-CD and the epoxy group of EPI in the alkaline environment. Under the catalysis of NaOH, the desired polymer was obtained by epoxy group opening and closing reactions.
The FT-IR spectra of β-CD, β-CDP, DS, and DS/β-CDP are shown in
Figure 1a. In the spectrum of β-CDP, the peak corresponding to the -OH had a significantly blue shift from 3341 cm
−1 to 3423 cm
−1 and the new peaks of C-O-C (1035 cm
−1 and 1084 cm
−1) appeared, indicating that the β-CDP had been successfully prepared through the ring opening and closing reactions. In the spectra of DS/β-CDP, an extra sharp band at 1738 cm
−1 could be assigned to the stretching vibration of C=O in the carboxyl group of DS, confirming that DS was successfully loaded into β-CDP.
Figure 1b shows the
1H NMR spectra of β-CDP, DS, and DS/β-CDP. Compared with β-CDP, the spectrum of DS/β-CDP showed new characteristic peaks at the range of 6.0–8.0 ppm derived from the protons of DS, indicating that DS had been successfully loaded in β-CDP. In addition, the UV–Vis absorption spectra of these samples are exhibited in
Figure 1c. After the loading of DS, a new absorption peak at 275 nm could be observed in the spectrum of DS/β-CDP, further indicating that the DS-loaded inclusion complex was prepared.
Since the solubility and activity (the active monomer quantity in β-CDP) of β-CD and β-CDP can affect the drug loading rate, the solubility and activity changes of β-CD before and after polymerization were detected and compared. As shown in
Figure 1d, the solubility of the β-CDP significantly increased to 0.855 g g
−1 in an aqueous solution at room temperature (25 °C), which was much higher than the solubility of β-CD (0.0184 g g
−1) [
20]. Then, the degree of polymerization was determined by the GPC by detecting the molecular weight distribution of β-CDP. The results showed that the retention time of β-CDP (Tr) was 20.13 min (
Figure 1e), and the average molecular weight and the degree of polymerization of β-CDP were calculated to be 5864 g mol
−1 and 5, respectively. According to the polymerization, the activity of β-CDP could still be maintained at 76.77% through the measurement of the PP probe technology (
Figure 1f). Finally, through calculations (Equation (3)), the drug loading capacity of β-CDP was higher than that of β-CD, which indicated that the influence of solubility was greater than the activity on the drug loading rate, so the β-CDP could effectively improve the solubility of the drug in water.
Figure 1.
Structure and properties of β-CDP. (a) FT-IR spectra of β-CD, β-CDP, DS, and DS/β-CDP. (b) 1H NMR spectra of β-CDP, DS, and DS/β-CDP. (c) UV–Vis spectra of β-CDP, DS, and DS/β-CDP. (d) Solubility and (e) GPC analysis of β-CDP. (f) UV–Vis spectra of PP/β-CDP and PP/β-CD with different content.
Figure 1.
Structure and properties of β-CDP. (a) FT-IR spectra of β-CD, β-CDP, DS, and DS/β-CDP. (b) 1H NMR spectra of β-CDP, DS, and DS/β-CDP. (c) UV–Vis spectra of β-CDP, DS, and DS/β-CDP. (d) Solubility and (e) GPC analysis of β-CDP. (f) UV–Vis spectra of PP/β-CDP and PP/β-CD with different content.
2.3. Preparation and Drug Release Behavior Analysis of DRL
Similarly, another L9 (3
4) orthogonal experiment including four principal factors (the dosage of PVA, β-CDP and DS, freeze–thaw cycle times) was conducted to obtain the desired DRL. The maximum cumulative release ratio of the model drug and the time of more than 85% release were considered as two test indexes to evaluate the sustained release ability of DRL. From the results (
Figure 3a,b, Equation (1), and
Table S2), the order of the main influencing factors for drug release was as follows: the dosage of DS > PVA > β-CDP > the freeze–thaw cycle times. When adding 1.5 g PVA, 1.0 g β-CDP, 0.2 g DS, and going through three freeze–thaw cycles (A
1B
1C
1D
3), the sample DRL exhibited the best mechanical strength (
Figure 3c,d). Then, the sustained release behavior of DRL with or without β-CDP was investigated (
Figure 3e). The DRL with β-CDP could prolong the drug release time, demonstrating that adding β-CDP could endow the hydrogel with the capacity to release the drug slowly.
Finally, we applied the release kinetics data of DRL to several typical drug release models (Equations (5)–(7)). We found the release mechanism of the DRL in PBS buffer solution (pH = 7.4) consistent with the Ritger–Peppas model (
Figure 3f). The rate constant “k” and diffusion index “n” were obtained from the above formula (
Table 1), and the diffusion index n was 0.14 (<0.45). The above results revealed that the drug releasing mechanism belonged to Fick diffusion [
21]. Because the DRL was prepared by physical crosslinking, the gel molecular chain relaxed quickly and failed to limit the release of the drug [
22,
23]. Hence, the drug diffused freely from the gel to the external environment by the concentration gradient. Apart from this, the drug in β-CDP would also be released slowly, prolonging the release time.
2.4. Preparation and Characterization of DL Hydrogel
We combined MSL and DRL to form a DL hydrogel dressing and then comprehensively investigated its mechanical properties, water retention capacity, skin adhesion property, and drug sustained release property.
The mechanical properties of the DL hydrogels were evaluated by detecting their rheological and tensile stress–strain curves. Storage modulus (G′, filled symbols) and loss modulus (G″, empty symbols) of hydrogels were measured as a function of angular frequency (0.1–100 rad/s) and strain sweep (0.0001–10%) (
Figure 4a,b). Over the entire frequency range, all G′ values of hydrogels were higher than the G″ values, meaning that all hydrogels were in the elastic solid gel state instead of the fluidic sol state [
24]. Obviously, the G′ of MSL was much higher than the G′ of DL and DRL, which meant the MSL could effectively provide enough mechanical strength for the DL hydrogel. Meanwhile, in the linear viscoelastic region of strain from 0.0001% to 0.01%, the G′ of the DL hydrogel was always much higher than others, owing to the formation of boron ester bonds in hydrogels. When the strain was more than 0.01%, values of G′ decreased dramatically, and the curves of each hydrogel had a crossover at a strain near 0.1%, which suggested that the gel network was totally broken and converted into a sol state at this strain [
25]. Next, the tensile stress–strain curves of the MSL, DL, and DRL hydrogels were investigated (
Figure 4c,d). The mechanical strength of MSL and DL hydrogel could reach 2892 kPa and 1504 kPa, respectively, while the mechanical strength of the DRL was only 12.81 kPa. Therefore, all these results demonstrated that the mechanical strength of DL hydrogel was strengthened by combining MSL with DRL [
26].
Materials with suitable water retention properties are essential in wound dressings because they provide a moist microenvironment to facilitate wound healing. Therefore, we conducted a series of contrast experiments to evaluate this property of our hydrogel (Equation (8),
Figure 4e,f). The results indicated that these hydrogels exhibited similar characteristics: on the one hand, for DRL, DL, and MSL, as the standing time reached 48 h, their water retention rate decreased from 100% to 46.05%, 67.31%, and 72.06%, respectively; on the other hand, when the water retention rate dropped to 70%, the DRL, DL, and MSL needed 5 h, 40 h, and more than 48 h, respectively. From the above two points, we can conclude that the introduction of MSL could effectively improve the water retention property of the DL hydrogel due to the existence of denser cross-linked networks for the reduction in water loss in the MSL.
As a dressing, excellent skin adhesion was also essential for hydrogels. As shown in
Figure 4g–i, the DL hydrogel could attach to the wrist, hand, and finger without falling, demonstrating that it had suitable adhesion ability due to the electrostatic and hydrogen bond interactions between the hydrogel and skin [
27].
Then, we detected the DS sustained release ability of DL hydrogel.
Figure 5a displays the drug release profile of our DS-loaded DL hydrogels. The DL hydrogel’s sustained release time could reach 8 h, and its total drug releasing amount was lower than DRL (59.1% < 93.4%). Unlike DRL, the drug release amount decreased sharply after 8 h and rose again at 20 h, forming an “S”-shaped curve and attracting our attention. As far as we know, the drug release rate of most typical hydrogels was constant when the cumulative release reached the maximum [
28,
29,
30,
31]. Therefore, we speculated that the drug release mechanism of the DL hydrogel was different from traditional hydrogel dressings.
To explore the drug release mechanism of DL hydrogel, we applied the release kinetics data of the DL hydrogel to several typical drug release models (
Figure 5b–d), and the fitting parameters are shown in
Table 2. Various mathematical models describing drug release provided insights into the mechanism of drug delivery. While diffusion was a prevailing mechanism for drug release from polymer networks, swelling and erosion might take over in some polymeric carriers. According to the results (
Table 2), Ritger–Peppas and Peppas–Sahlin mathematical models indicated diffusion and diffusion–relaxation controlled systems, which failed to describe DS release from the DL hydrogel. Based on the drug release mechanism of DRL, the m in the diffusion–relaxation–erosion model was determined as 0.5 from
Table 3 [
32]
1 2. Compared to other models, this semi-empirical equation’s corresponding regression coefficient (R
2 = 0.9902) was the closest to 1, so we concluded that the drug release behavior of the DL hydrogel obeyed the diffusion–relaxation–erosion model. The reasons for this phenomenon can be summarized as follows: compared with the drug release of monolayer hydrogel, the DL hydrogel not only released drugs directly to the external solution but also diffused part of drugs from the DRL to the MSL, and then released them to the solution. Therefore, the DL hydrogel’s release time would be prolonged, but the release rate would decrease. At the same time, because of the swelling phenomenon caused by the double-layer structure of DL hydrogel, it could lead to a specific drug absorption. Then, with the prolongation of the soaking time, the DRL of the hydrogel would be corroded to release more drugs and the MSL would be swollen, so that some of the reabsorbed drugs would release again [
22].
Finally, we compared our DL hydrogel dressing with other dressings (such as layered dressing and hydrogel microspheres, etc.) reported in the field of drug delivery in recent years (
Table 4). It could be concluded that traditional hydrogels had poor performance in mechanical strength and water retention, especially in the field of hydrogel dressing. Unlike previous works, we first applied the DL structure to manufacturing hydrogel dressing to simplify the design of a multifunctional single-layer dressing, endowing the hydrogel with better mechanical strength (G′ = 2.5 × 104 Pa) as well as water retention (48 h, 67%). This structure could prolong the drug release process simultaneously. Benefiting from the DL structure, this drug release model of hydrogel was different from previous work (diffusion–relaxation–erosion model rather than the Ritger–Peppa model). In summary, our design could equip hydrogel dressings with many excellent properties and broaden the practical application range.