The drugs 5-FU and Reglan were purchased from TCI Shanghai Co., Ltd. (Shanghai, China). The polymeric matrices PVP K30 (M _w = 58,000) and CA (M _w = 50,000) were purchased from BASF Shanghai Co., Ltd. (Shanghai, China) and Shanghai Haosheng Biotechnol. Co., Ltd. (Shanghai, China), respectively. The organic solvents anhydrous ethanol, chloroform (analytical grade), and dimethylacetamide (DMAc) were obtained from Sinopharm Reagent Co., Ltd. (Shanghai, China). All the other raw materials were of analytical grade, and water was double-distilled before use.

After some preliminary experiments, four working fluids were determined for the combined fabrication process: 1) For preparing the monolithic 5-FU/CA composite particles, 3.0 g of 5-FU and 9.0 g of CA were co-dissolved in 300 mL of a solution containing DMAc, acetone, and ethanol in a volume ratio of 2:3:1 correspondingly. 2) To prepare the core-shell particles via coaxial electrospraying, the core fluid was prepared by dissolving 6.0 g of 5-FU and 9.0 g of CA in 300 mL of a solution containing DMAc, acetone, and ethanol in a volume ratio of 2:3:1; further, the shell fluid was prepared by dissolving 9.0 g of CA in 300 mL of a solution containing DMAc, acetone, and ethanol in a volume ratio of 1:4:1. 3) For the casting film preparation, 3.0 g of Reglan and 15.0 g of PVP K30 were dissolved in 100 mL of chloroform, and 12.0 g of 5-FU-loaded core-shell particles were suspended in this Reglan/PVP solution.

A homemade concentric spinneret was used to set up the coaxial electrospraying apparatus. The other parts included two fluid drivers (KDS100 and KDS200, Cole-Parmer, United States of America) for quantitatively pumping the core and shell working fluids with an accuracy of 0.01 mL/h, a high-voltage generator (60 kV/2 mA, Wuhan Hua-Tian High Power Co., Ltd., Wuhan, China), and a collector comprising a cardboard wrapped with aluminum foil. After some optimizations, the working conditions were determined as follows: core and shell fluid flowrates of 1.0 and 1.0 mL/h, respectively; an applied voltage of 18 kV; and a collection distance of 20 cm between the nozzle of the spinneret and collector. The suspensions containing Reglan and core-shell particles were degassed through an ultrasonic instrument in an ice bath for better implementation (Wang et al., 2023a). Later, the suspensions were placed in an oven at a temperature of 40°C until a constant weight was achieved. Other parameters are included in Table 1.

The drug loading efficiencies (LEs, %) were calculated according to Eq. (1): L E % = Q d Q p × 100 % (1) where Q _d and Q _p represent the detected and theoretical amounts of the drugs during preparation. To prepare the electrosprayed medicated particles S2 and S3, specific amounts of their powders were weighed and dissolved in a solution containing DMAc, acetone, and ethanol in a volume ratio of 2:3:1. Then, approximately 1 mL of each solution was dripped into 100 mL of water under ultrasonic conditions. After filtration, the aqueous solution was measured to determine the 5-FU loading in the particles. The LE (%) value of Reglan in the solid film was determined through an in vitro dissolution test.

The in vitro dissolution tests were conducted in accordance with the Chinese Pharmacopoeia (2020 Ed.), and the paddle method was employed along with an RCZ-8A dissolution apparatus (Tianjin University Radio Factory, China) and seven vessels. The test conditions involved a rotation speed of 50 rpm and a dissolution media temperature of 37°C ± 1°C. For the medicated products S2, S3, and S4, powders of respective weights of 0.1 g, 0.1 g, and 0.25 g were placed in the vessels with a constant 5-FU feed. The dissolution media (600 mL) used was 0.01 N HCl (pH = 2.0) for the first 2 h to mimic the gastric juices artificially, and an equivalent volume of sodium hydroxide was subsequently added to the dissolution media to adjust the pH to 7.0 to simulate the intestinal fluids artificially. For the S4 sample, the residues from the in vitro dissolution test were separated and dried naturally for SEM observations.

At predetermined time intervals, 5.0 mL volumes of the dissolution media were withdrawn for sampling, and equal volumes of fresh media were added to maintain constant volumes. The absorbance values of the samples were measured using a UV‒vis spectrophotometer (Unico Instrument Co., Ltd., Shanghai, China). The amounts of Reglan and 5-FU in the samples were then calculated using their predetermined calibration curves. No mutual interference was observed between the dissolutions of Reglan and 5-FU due to the blank CA coating in the core-shell particles.

All the experimental data are presented as mean ± SD. The results from the in vitro dissolution tests were analyzed using a one-way ANOVA, and the significance level threshold was set at 0.05. Thus, p (probability) values lower than 0.05 were considered to be statistically significant.

Electrospinning as an advanced nanofabrication technique is rapidly gaining applications in a wide variety of fields (Bai et al., 2022; Cao et al., 2022; Chen et al., 2023a; Shi et al., 2024b); its success is attributable to a series of unique properties, such as straightforward, single-step, and flexible fabrication, in addition to the use of an inexpensive apparatus (Sivan et al., 2022b; Lv et al., 2024a; Lv et al., 2024b). Among these, its flexibility allows conveniently combining electrospinning with many traditional techniques and advanced chemical and physical methods (Yu and Xu, 2023; Yu and Zhou, 2024). Similar to electrospinning, electrospraying also allows facile combinations with other fabrication methods to develop new material conversion approaches. A diagram showing the compatibility between coaxial electrospraying and solvent casting is depicted in Figure 1.

Effective conversion of the electrosprayed medicated particles to suitable dosage forms is always an important process in drug delivery (Alfatama et al., 2024). After preparation, the core-shell particles can be placed in capsules for direct oral administration or subjected to additional processing to add new types of functions. Solvent casting is a common method for creating lipid and polymeric films in pharmaceuticals. The further casting of the electrosprayed core-shell particles is useful not only for convenient oral delivery and easy shipping and storage but also for loading new active ingredients for synergistic therapy. This proof-of-concept idea is expected to be useful for developing other multifunctional biomaterials in the future.

It must be noted that coaxial electrospraying is the key technique in the proposed combination. In this study, a detachable spinneret was developed for the coaxial electrospraying process, which has also been used in other EHDA processes, such as single-fluid electrospinning and coaxial electrospinning (Chen et al., 2023b; Zheng et al., 2024). The Teflon-coated concentric spinneret consists of a common metal-based concentric spinneret and a set of Teflon tubings (Figure 2a1). The Teflon tubing can be moved vertically to adjust the nozzle surfaces of the inner and outer capillaries, which allows further adjustments of the behaviors of the shell and core working fluids (Figure 2a2).

The complete homemade electrospraying system is shown in Figure 2B and comprises two pumps, a collector, a spinneret, and a power supply (Alfatama et al., 2024). Under optimized experimental conditions, the typical process for generating the core-shell particles S3 is as shown in Figure 2B; the inset image on the bottom left shows the deposition of the electrosprayed particles over the letters “ZNU” formed by covering the surface of the collector with a sheet of paper for approximately 10 min. During the process, the area around the spinneret that is connected to the two working fluids is considered important, along with the alligator clip connected to the power supply (Figure 2C). All the electrospraying processes are initiated at this point of convergence upon the reasonable formation of a stable Taylor cone (Ji et al., 2023; Xu et al., 2023).

To prepare the homogeneous 5-FU/CA composite particles S2, only the core metal capillary was used to guide the working fluid toward the nozzle of the spinneret. After applying a high voltage of 18 kV, the atomization can be discerned, as shown in Figure 2D. The atomization process consisted of three successive steps, i.e., Taylor cone (Figure 2E) formation, a straight fluid jet (often condensed to a point at the tip of the Taylor cone), and Coulombic explosion region formation. When coaxial electrospraying was performed, the Teflon tubing was moved downward to allow outward projection of only 0.2 mm of the core metal capillary. This arrangement was favorable for fine encapsulation of the core fluid by the shell fluid. The atomization is recorded in Figure 2F. The working process here is apparently similar to the single-fluid process comprising the Taylor cone, joint point, and successive Coulombic explosion sections. However, the different compounds of the Taylor cone can be discerned by further enlarging the observations, as shown in Figure 2G.

The morphologies and inner structures of the electrosprayed particles and hybrid films are shown in Figure 3. As anticipated, the CA particles S1 electrosprayed from the blank polymeric solution had a somewhat concave morphology, as shown in Figure 3A. Meanwhile, there are many satellites around the electrosprayed particles. In contrast, the 5-FU/CA particles S2 had a spherical morphology, as indicated in Figure 3B. Similarly, the core-shell particles S3 from the coaxial electrospraying process had round shapes, as exhibited in Figure 3C and in the inset image on the top right at a magnification of ×50,000. In general, the hybrid films were brittle, and cracks were observed after being manually broken by external forces, as shown in Figure 3D. The SEM morphologies of the hybrid films are shown in Figures 3E, F at various magnifications. From Figure 3F, it is clear that the surrounding regions are rougher than the central sections of the particles, indicating that each particle had a double-compartment core-shell structure.

Electrosprayed particles often have a broader diameter distribution than electrospun nanofibers; the particle sizes and morphologies are influenced by a series of solution properties and operational parameters. In this study, pure CA particles S1 (Figure 3G), 5-FU/CA composite particles S2 (Figure 3H), and core-shell particles S3 (Figure 3I) had average diameters of 1.07 ± 0.21 μm, 0.87 ± 0.14 μm, and 0.78 ± 0.11 μm, respectively. It is interesting to note that the composite particles S2 and the core-shell particles S3 had smaller diameters and more uniform size distributions than the blank CA particles S1, even though high drug loading was performed within S2 and S3.

TEM images were obtained to evaluate the inner structures of the composite particles S2 and core-shell particles S3. As seen in Figure 4A, the 5-FU/CA particles S2 had homogeneous structures. Under a bright field, the gray levels of the TEM images are a result of thickness, density, and elements. The gradually decreasing gray levels in Figure 4A are attributed to the thicknesses of the particles, with the darkest level at the center and the lightest levels near the boundaries. Overall, the elements and densities were similar throughout the particles. In contrast, the core-shell particles S3 had stepwise gray level changes, as indicated by the dashed circular line in Figure 4B. The reasons for these stepwise changes include different elements (the shell had no 5-FU), different thicknesses, and different densities.

The microformation mechanisms of the blank CA particles S1 and composite 5-FU/CA particles are compared in Figure 5. In general, the solidification processes of electrospinning and electrospraying are very different; the former is mainly a fluid drawing process, whereas the latter is a continuous fission process. From the Taylor cone to the collector, the working fluid experiences numerous fission reactions; each fission reaction inevitably results in a downsizing of the fluid droplets, a decrease in the solvent amount and related increases in the polymer concentrations within the droplets, a weakening of the surface charges and splitting forces, and proximity to the collector. The fission reactions continue until the repelling forces cannot split the droplets/solid particles/semisolid particles further.

Thus, we divide the entire Coulombic process into four subprocesses, as indicated in Figure 5. These four steps are: i) quick splitting; ii) slow splitting; iii) semisolid shell formation; and iv) particle solidification. The most common characteristics of each step are as follows. The quick splitting occurs after Taylor core formation when the strongest repelling forces atomize the fluid with the lowest solute concentration. Next, the slow splitting continues with the fluid droplets that have smaller repelling forces but high solute concentrations. Here, the surface and inner sections still have homogeneous components and compositions. When the droplets further approach the collector, a semisolid surface forms on the split droplet, whose surface and inner sections have heterogeneous components and compositions. Finally, the solidified particles from the semisolid surface droplets may still contain some solvents in their inner sections; owing to the presence of these residual solvents, the solidified particles may experience a barometric pressure and are flattened to concave morphologies. This is the mechanism of formation of the blank CA particles S1.

However, the addition of a large amount of 5-FU to the CA fluid can exert a remarkable influence on the entire electrospraying process. First, these small molecules may retard the formation of semisolid substances and compact membranes on the surfaces of the split droplets, thereby facilitating the fast removal of the solvent molecules from the inner sections of droplets. Second, when the collected particles experience a barometric pressure owing to the evaporation of the residual solvent molecules, the loaded molecules act as supports between the macromolecular chains to prevent deformation. This is the mechanism of formation of the spherical 5-FU/CA composite particles S2. As for the core-shell particles S3, the diluted shell CA solution may further facilitate the removal of the inner residual solvent molecules, thereby further guaranteeing a spherical morphology.

For drug delivery, particularly for drugs with poor water solubilities, the amorphous state is more favorable than the crystalline state for predictable, controlled release (Pattnaik et al., 2023; Sun et al., 2024). The XRD patterns of the four starting components (Reglan, 5-FU, PVP, and CA) and their homogeneous composite S2, heterogeneous composite S3, and hybrid casting films S4 are shown in Figure 6. As indicated by the abundant sharp Bragg peaks in the patterns, both the raw 5-FU and Reglan powders are crystalline components, whereas the polymeric matrices CA and PVP are amorphous substances. When the working fluids are processed through single-fluid electrospraying, coaxial electrospraying, and solvent casting, all the converted solid products, i.e., monoaxial particles S2, core-shell particles S3, and hybrid films S4, are obtained in the desired amorphous state. The reasons for this are mainly attributed to the drug distribution on the molecular scale within the polymeric matrices, which resulted from the homogeneous working fluids, the extremely fast drying effect of electrospraying, and the fine compatibility between the drug and polymer molecules.

Good compatibility between a drug and its carrier is important for realizing not only the desired controlled release profile of the drug but also chemical and physical stability for storage and shipping. New types of polymers are continuously being tested for potential biomedical applications (Guo et al., 2021; Köse et al., 2022; Liao et al., 2023), and their compatibility with the loaded drugs is being disclosed (Ejeta et al., 2022; Jurić et al., 2023; Langer and Peppas, 2024; Song et al., 2024). FTIR spectrometry is one of the popular methods used to assess the compatibilities of components. In this study, the ATR-FTIR spectra of the four starting components, i.e., Reglan, 5-FU, PVP, and CA (left), and their molecular formulas (right), are shown in Figure 7. Comparing the spectra of the drugs with those of the loaded products, it is obvious that there are many observable sharp peaks for the raw drug powders; however, all of these were remarkably reduced or even attenuated in the spectra of the final products. This phenomenon suggests that the drug molecules interact with the polymeric molecules through secondary physical interactions, such as hydrogen bonding, hydrophobic interactions, or electrostatic interactions. Although these interactions cannot be directly detected using instruments, they can be deduced through their molecular formulas.

The LE (%) values for 5-FU in the electrosprayed particles S2 and S3 were 99.26% ± 2.14% and 100.3% ± 1.77%, respectively. The electrospraying processes, regardless of the single-fluid blend process for creating homogeneous S2 or the coaxial process for producing core-shell S3, are essentially physical drying processes that are completed very rapidly. Thus, the drug 5-FU, without sublimation or volatilization properties, can be completely encapsulated in the solid particles.

The in vitro drug release profiles of Reglan and 5-FU from their host polymeric matrices are shown in Figure 8. The antiemetic drug Reglan is released rapidly when the hybrid films are placed in the dissolution media, as indicated in Figure 8A. Meanwhile, the detected total release content of Reglan reached 99.47% ± 3.57% of the theoretical content calculated from the film casting suspensions. The pulsatile release of Reglan is highly desirable for patients because it allows for rapid initiation of antiemetic actions. As for 5-FU release from the corresponding products, i.e., 5-FU/CA composite particles S2, core-shell medicated particles S3, and particles released from the hybrid films S4, the profiles are compared in Figure 8B. The 5-FU release profiles over the initial 4 h are shown in Figure 8C.

As anticipated, 5-FU was released from the medicated materials in a sustained manner over a longer period of time. However, the sustained release details were significantly different in terms of the initial manner of release and sustained drug release rates. The composite particles S2 released 33.4% ± 4.3% of the loaded 5-FU in the first hour, showing a typical initial burst-release effect. After 6 h, approximately 70.6% ± 4.5% of the loaded 5-FU was freed into the dissolution media. In sharp contrast, the core-shell particles S3 and the particles from the hybrid films S4 released no detectable 5-FU during the first half hour, and only 2.7% ± 0.9% and 3.4% ± 1.1% of the 5-FU were released over the first 2 h, respectively. The blank CA shell layer played an important role in retarding the initial 5-FU release. The release amounts over the first 6 h for S3 and S4 were 30.2% ± 3.9% and 29.8% ± 3.9%, respectively. Later, 5-FU showed sustained release in an almost linear manner up to 30 h. These release profiles of the 5-FU/CA core-shell particles thus ensure the desired asynchronous dual-drug delivery from the hybrid films S4.

The regression equations of the drug 5-FU released from the medicated products are shown in Figure 9 based on the in vitro dissolution test data. For the three medicated products, i.e., homogeneous 5-FU/CA particles S2, core-shell 5-FU/CA particles S3, and core-shell particles released from the hybrid films S4, the release equations are LogQ₂ = 1.536 + 0.323 Logt (R = 0.9593), LogQ₃ = 0.282 + 1.240 Logt (R = 0.9607), and LogQ₄ = 0.429 + 1.120 Logt (R = 0.9717), respectively. Based on the critical value of the Peppas equation (Peppas, 1985), the value of the sample S2 (0.323) is less than 0.45, indicating that 5-FU was released from the CA-based particulate composites in a typical Fickian mechanism. Furthermore, the values of the electrosprayed core-shell particles S3 and the particles freed from the hybrid films S4 were n = 1.24 and 1.12, respectively. A value exceeding 0.9 suggests an erosion mechanism based on the Peppas criterion. However, since CA is an insoluble polymer, it is determined that 5-FU was released by a two-step penetration process; the first step involved the penetration of water molecules into the interior of the CA matrix. Later, the second penetration step involved 5-FU molecules passing through the inner channels formed by the water molecules into the bulk dissolution media. This contradiction is unexpected and is attributed to the blank CA coating on the 5-FU/CA composite in the core-shell structures. The Peppas equation is useful for evaluating drug-loaded materials in which the drug molecules are distributed homogeneously throughout the polymeric matrices, but it fails to predict nanomaterials with complicated chamber structures, as has been demonstrated in other investigations (Zhou et al., 2023; Zhao et al., 2024; Zhou et al., 2024a; Wang et al., 2022).

This study demonstrates a proof-of-concept strategy for creating synergistic anticancer DDSs based on a combination of the traditional film casting technique and advanced coaxial electrospraying. Knowledge of the working mechanism is generally useful for developing a series of functionally medicated nanomaterials. A diagram and two digital photographs of the casting films are shown in Figure 10. The hybrid film is semitransparent owing to the presence of numerous electrosprayed core-shell particles, as indicated by the red letters “ZNU” in Figure 10A. The film had a thickness of approximately 2.0 mm with a smooth cross-section, reflecting its fragility, which could be improved simply by employing PVP with a higher molecular weight (such as PVP K90) but sacrificing a little of the fast dissolution property.

The asynchronous dual-drug release profiles are shown in detail in Figure 10C. First, the fast dissolution of the soluble PVP film matrix releases the therapeutic ingredient Reglan to achieve the desired antiemetic application in a pulsatile manner. Then, the loaded electrosprayed core-shell particles are simultaneously released into the dissolution media; because of the insoluble CA shell coating on the core drug/polymer medicated composites, there is no initial drug release. Later, the penetration of water molecules into the core-shell particles causes a slight swelling of the particles. In turn, this creates passages for material transportation between the solid particles and the dissolution media for sustained release of the encapsulated 5-FU molecules. Because of the high amount of drug loading in the core section, the CA particles are flattened after 5-FU depletion. This can be deduced from the SEM images of the residue particles (Figure 10D), which were fetched from the dissolution vessels and dried naturally.