Glacial acetic acid (CH₃COOH), tetrabutyl titanate (TBOT), absolute ethyl alcohol (C₂H₅OH), urea (CO(NH₂)₂), nitric acid (HNO₃), deionized water (DI), dichloromethane (DCM), dimethyl formamide (DMF) and cerium (III) nitrate hexahydrate (Ce(NO₃)₃·6H₂O), and diphenylisobenzofuran (DPBF) were purchased from Aladdin Shanghai. Polylactic acid granules (4032D) were purchased from NatureWorks (USA). Luria–Bertani (LB) and Tryptic Soy Broth (TSB) culture media, agar powder, were procured from Haibo-bio (China).

Nitrogen-doped titanium dioxide (N-TiO₂) and a series of cerium nitrogen co-doped TiO₂ (Ce-N-TiO₂) were prepared according to the methods of Meng et al. (2014) and Nasir et al. (2014). In total, 30 mL of ethanol and 2 mL of ultrapure water were taken into a 250-mL round-bottom flask fitted with a mechanical stirrer; 2 g of urea was used as nitrogen precursor, and varied amounts (0.05, 0.1, 0.2 g) of cerium nitrate hexahydrate (Ce(NO₃)₃6H₂O) were totally added into the above solution to form “solution A.” In total, 30 mL of ethanol and 3 mL of 5 mol/L nitric acid solutions were added into another flask; 10 mL of TBOT was slowly added to this solution with continues stirring for 0.5 h to form “solution B.” Finally, solution B was added slowly into solution A at the rate of 1 mL/min under moderate stirring. Stirring was continued for another 2 h and then transferred into a 100-mL Teflon-lined autoclave, which was kept in the oven at 120°C for 16 h. After the autoclave was cooled to room temperature, the precipitates at the bottom of the autoclave were washed with deionized water several times to remove the impurities. The obtained samples were kept in the oven at 100°C overnight for drying. In order to have better crystallization and to remove the solvent and other impurities, the samples were calcined at 480°C for 3 h.

The obtained photocatalysts were denoted as N-TiO₂, 0.05Ce-N-TiO₂, 0.1Ce-N-TiO₂, and 0.2Ce-N-TiO₂, respectively; 0.05, 0.1, and 0.2 represent the grams of cerium nitrate added.

PLA solution (8 wt%) was prepared by the addition of PLA particles into 8:2 (m: m) DCM/DMF mixture solvents and magnetic stirring for 20 h until well combined. Then, 5 wt% TiO₂ NPs were dispersed in a DCM/CMF solvent through sonication for 30 min. The applied working distance (distance between the needle tip and collector), positive voltage, and flow rate were 20 cm, 20 kV, and 1 mL/h, respectively. The electrospinning environment was controlled at 25°C and 35% humidity, and the collection time was 4 h. The corresponding electrospun PLA membranes were PLA electrospun without TiO₂ (PLA) and PLA electrospun loaded TiO₂ (PLA-N, PLA-0.05, PLA-0.1, PLA-0.2). The PLA was vacuum dried at 60°C for 6 h to remove residual solvent.

The sample morphologies were characterized using a scanning electron microscope (SEM, JSM-6390LV, Japan) and a transmission electron microscope (TEM, JEM-2100Plus, Japan). Dynamic light scattering (DLS) measurements were performed by Zetasizer Nano ZS (Malvern, UK) equipped with a He-Ne laser (633 nm, 4 mW) with deionized water as the dispersant. The crystal structure of TiO₂ was identified using an X-ray diffractometer (XRD, Smart Lab, Japan) with Cu-kα radiation (λ = 1.5406 Å). X-ray photoelectron spectroscopy (XPS) analysis was utilized to determine the surface chemical species of the particles (Escalab Xi+, Thermo Fisher, Czech Republic). The UV–visible diffuse reflectance spectra in the range of 200-800 nm were measured using a spectral electrochemical instrument (Autolab, Switzerland), with BaSO₄ serving as the reflectance standard. The chemical bonds and other characteristics of the samples were analyzed using Fourier transform infrared spectroscopy (Nicolet iS50, Thermo Fisher, USA) to provide a comprehensive understanding of their properties.

The 1,3-diphenylisobenzofuran (DPBF) exhibits high specificity toward singlet oxygen (¹O₂), and after binding with singlet oxygen (¹O₂), DPBF will be irreversibly oxidized, and the absorption intensity at 410 nm will rapidly decrease. This enables the detection of ROS generation; 5.0 mL solution of acetonitrile containing DPBF (10 μg mL⁻¹) was mixed with PLA, PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 electrospun samples (1.5 × 1.5 cm²). The DPBF solution (mixed with PLA electrospun) was then placed in a camera bellows. The PLA electrospun was irradiated with visible light from a 2mW cm⁻² xenon lamp equipped with a UV filter for 30 minutes. The UV–Vis spectrophotometer (UV-2700, Japan) was utilized to measure the absorbance values of each solution at 415 nm during different time intervals.

The ROS Assay Kit (Beyotime, Shanghai) was used to measure the ROS in L929 cells treated with PLA, PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 electrospun samples. In short, L929 cells were seeded into six-well plates at 4 × 10⁴ cells per well and divided into seven groups: the first group was the blank control group; the second group was the positive control using Rosup reagent in the ROS Assay Kit. The medium containing nanoparticles was then replaced with serum-free medium supplemented with 10 uM DCFH-DA, and incubated at 37°C for 30 min. The cells were washed three times with serum-free medium and placed under an inverted fluorescence microscope for observation.

The contact angle of water was characterized by using a contact angle analyzer (Krüss DSA30S, Germany). Distilled water (2 μL) was deposited onto each test surface to measure the contact angle.

Briefly, the PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 membranes were initially weighed (dry weight) before immersion in PBS (pH 7.4) at 37°C for 2 and 24 h. The weights were specified as m₀ before immersion and m₁ after immersion. Therefore, the swelling ratio formula is swelling ratio = (m₁–m₀)/m₀ × 100%.

Staphylococcus aureus (S. aureus, ATCC 6538) and Escherichia coli (E. coli, ATCC 8739) were purchased from Solarbio (Beijing) and selected as representative strains of Gram-positive (G+) and Gram-negative (G-) bacteria. Colony-forming unit (CFU) assay was used to assess the antibacterial properties of several PLA materials against the two bacteria. After being cultured for 24 h on the Tryptic Soy Broth (TSB) agar, the S. aureus colony was isolated and inoculated into a TSB fluid medium and shaking in a bacterial shaker speed of 150 rmp rotation at 37°C for 18 h. Similarly, after being cultured for 24 h on Luria–Bertani (LB) agar, the E. coli colony was carefully selected and inoculated into Luria–Bertani medium and shaken in a shaker at a rotation speed of 150 rpm rotation at 37°C for 18 h. Then, these two different bacterial suspensions were subsequently diluted to 10⁶ CFU/mL; 100 μL of the prepared bacterial suspension and PLA, PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 membranes (0.6 × 0.6 cm) were mixed in a 96-well plate. The antibacterial experiments were carried out under two conditions: one was a visible light source (2 mW cm⁻² xenon lamp with a UV filter) irradiation for 30 min, which we called “Light”; and the other was without irradiation, which we called “Dark,” so as to compare and analyze the influence of visible light excitation Ce-N-TiO₂ on bactericidal effect. After 30 min of dark treatment and visible light irradiation conditions, 50 μL of the bacterial suspension was taken out and diluted to 2 mL before being cultured in a shaker for 2 h. Then, the bacterial suspensions were, respectively, inoculated on TSB or LB agar and incubated at 37°C for 24 h. PLA was tested as a control (Wang et al., 2022). Colony-forming units (CFU) are counted, and the antibacterial efficiency of each plate is calculated as follows:

where A = CFU in the control group or CFU in the experiment group, and B = counts of colonies in the control group.

In order to observe the characteristics of bacteria growth and antibacterial effect more intuitively, bacteria growth curve tests were introduced into experiments. After undergoing the same experimental treatment, the bacterial suspension was divided into groups and cultured in individual wells of a 96-well plate at 37°C. PLA was also tested as a control. The absorbance change was measured every 2 h for 12 h using a microplate reader set to detect at 630 nm.

In this experiment, the cytotoxicity of PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 membranes was assessed through the CCK8 assay and calcein AM/PI live/dead kit (Beyotime, China). The L929 murine fibroblast cell culture medium was composed of Dulbecco’s Modified Eagle Medium (DMEM) with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin solution. In brief, the L929 cells cultured in flasks were digested using 0.25% trypsin–EDTA solution, resuspended in the medium for cell counting, and then, 200 μL cell suspension was seeded at a density of 1 × 10⁴ cells per well in 96-well plates incubated at 37°C with 5% CO₂ for 24 h. Then, the 0.6 × 0.6 cm² PLA electrospun membrane was carefully positioned onto the base of a 96-well plate. An experimental group was placed in a dark incubator, while another experimental group was exposed to light for 30 min before being placed in a dark incubator. The control group experiments without PLA membranes were executed in the same procedure as described before. After 18 h of incubation, the original culture medium in each well was removed, and 200 μL of fresh culture medium and 20 μL of CCK8 solution were added and incubated for an additional 2 h. The OD value at 450 nm was measured by a microplate reader (Molecular Devices, LLC). The cytotoxicity of the PLA membrane was calculated from the formula:

where As = experimental wells, Ab = blank wells, and Ac = control wells.

As the cell culture procedure described previously, the medium was removed and washed twice with PBS, and 100 μL calcein AM/PI was added to each well before incubation for 30 min in the dark. After incubation, the staining effect was observed under a fluorescence microscope (calcein AM was green fluorescence, Ex/Em = 494/517 nm; PI is red fluorescence, Ex/Em = 535/617 nm). Green fluorescence represents living cells and red fluorescence for dead cells.

The experiment was repeated at least three times. Experimental data were presented as mean ± standard deviation (SD) (n ≥ 3). Significant differences between data were tested by analysis of variance (ANOVA). In all evaluations, p < 0.05 was considered as statistically significant; p < 0.05, p < 0.01, and p < 0.001 were denoted by (*), (**), and (***), respectively.

The membrane surface hydrophobicity–hydrophilicity properties were investigated by water contact angle analysis. As shown in Figures 4A–E, most of the membranes exhibit hydrophobic properties with contact angles ranging from 140° to 125°. With the addition of TiO₂ NPs, the contact angles of PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 films declined to a certain extent, indicating that the water wettability of PLA membranes was improved.

ROS plays a pivotal role in photodynamic antibacterial therapy (PDT), photocatalytic antibacterial therapy (PCT), and sonodynamic antibacterial therapy (SDT) modalities. Sufficient ROS, including superoxide anion radical (O₂^·−), hydrogen peroxide (H₂O₂), singlet oxygen (¹O₂), and hydroxyl radical (·OH), can impair bacterial membrane phospholipids and membrane proteins, leading to the disruption of bacterial structure integrity, intracellular substance leakage, functional inactivation of the membrane transport system, and related protease. Moreover, ROS can irreversibly damage the components of DNA, thereby causing double-strand breaks and inhibiting the growth and reproduction of bacteria (Jia et al., 2019; Feng et al., 2021).

The fluorescent probe DCFH-DA (2,7′-dichlorodihydrofluorescein diacetate) is widely employed for intracellular ROS testing due to its good lipophilicity and membrane permeability. Upon oxidation by ROS, DCFH yields 2′,7′-dichlorofluorescein (DCF), which exhibits intense fluorescence that can be visualized using fluorescence microscope.

Figures 4F–L show the images of intracellular ROS staining; Figures 4F,G show the negative control group and positive control group, respectively. Figures 4H–L exhibit a gradual increase in fluorescence intensity, indicating elevated intracellular ROS levels under visible light treatment with PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2. Notably, PLA-0.2 displayed the highest fluorescence intensity, suggesting that PLA-0.2 enhanced intracellular ROS the most and was more conducive to playing a bactericidal role (Mickymaray et al., 2023).

1, 3-diphenylisobenzofuran (DPBF) is a highly specific fluorescent probe for singlet oxygen (¹O₂), which serves as an indicator of its presence. As shown in Figure 4M, after 30 min of exposure to light, the PLA group did not exhibit a significant decrease in absorbance at 411 nm. In the other four groups, the absorbance of DPBF solution at 411 nm exhibited a continuous decrease with increasing light exposure time, particularly in PLA-0.2. These results proved that the PLA membranes loaded with Ce-N-TiO₂ increased their photocatalytic activity and produced more ROS under visible light conditions as the proportion of Ce doped with TiO₂ increased.

For medical biological dressings, the evaluation system must include properties such as water absorption and swelling ratio, which are crucial for creating a favorable wound microenvironment and promoting healing (Hajikhani et al., 2021). The maintenance of wound cleanliness and infection control is essential throughout all stages of wound healing. Additionally, a moist wound that is not too dry or wet plays a crucial role in wound treatment in order to accelerate the healing process (Powers et al., 2016). In addition, the normal exudation of tissue fluid or blood from the wound, along with the inflammatory secretion from an infected wound, significantly impacts the microenvironment of the wound and ultimately influences the entire process of wound recovery. Therefore, the dressings used to cover the wound surface should possess excellent water absorption and moisturizing properties, in addition to antibacterial activity and good biocompatibility (Liang et al., 2021; Peng et al., 2022; Zeng et al., 2022).

The 2 h and 24-h swelling ratio performance is shown in Figures 4N,O. After immersion in PBS for 2 h, the water absorption of the pure PLA electrospun membranes prepared according to the parameters described above was close to 160%, while the rate of the PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 electrospinning membranes significantly increased to different degrees. In addition, after immersion in PBS for 24 h, the swelling ratio of PLA, PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 was approximately 220%. Among them, the ratio of PLA-0.05 was slightly higher, approximately 260%. It was plausible that the hydrophilic properties of TiO₂ NPs within PLA may be accountable. This also could be attributed to TiO₂, which can easily penetrate the unsaturated bonds surrounding the PLA polymer chain, thereby enhancing its capillary effect and increasing water absorption through this effect. Simultaneously, changes in diameter distribution and void ratio of the electrospun membrane may also improve the swelling ratio.

The antibacterial dressings containing photosensitizers capable of generating ROS upon irradiation were employed in antibacterial photodynamic therapy. ROS are molecules that gain electrons and are reduced to free radical forms, including superoxide anion O₂^·−, peroxide O₂^·−, hydrogen peroxide H₂O₂, hydroxyl radical OH, and hydroxyl OH− ions (Dryden, 2018). The primary mechanism by which ROS eliminates bacteria is through the disruption of bacterial cell walls and internal structure as well as the interference with normal physiological activities. In the typical photocatalytic antibacterial model, ROS can inflict structural damage on the bacterial membrane and induce perforation, thereby impairing its semi-permeability and resulting in the loss of bacterial contents. Furthermore, ROS can also induce damage to the DNA, mRNA, ribosomes, and essential proteins in bacteria (Pelgrift and Friedman, 2013; Thakur et al., 2019; Huo et al., 2021). Various unique mechanisms endow ROS with exceptional efficiency and superiority. This disruption leads to an imbalance in material exchange and ultimately leads to bacteria death. It is worth noting that metal oxides such as ZnO and TiO₂ NPs also have a bactericidal effect after direct contact with bacteria under non-light conditions, which may be related to the positive charge on the surface of the nanoparticles (Zhang et al., 2021). In order to confirm the antibacterial efficacy of various PLA membranes, CFU reduction and absorbance curves were employed to assess their antimicrobial properties. S. aureus and E. coli were, respectively, employed as the representative strains of Gram-positive and Gram-negative bacteria to evaluate the antibacterial efficacy of PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2. Pure PLA was set as the blank control. As shown in Figure 5, the inhibitory effect of PLA and PLA-0.05 on bacterial growth was found to be negligible. Furthermore, the antibacterial efficacy against Gram-positive (S. aureus) bacteria was more pronounced compared to that against Gram-negative (E. coli), especially in PLA-0.1 and PLA-0.2.

The survival rate changes of S. aureus are shown in Figures 5A,C. After 30 min of irradiation, all groups except PLA-N were effective in eliminating S. aureus. The S. aureus average survival rate of PLA-0.05, PLA-0.1, and PLA-0.2 was 28.29, 6.98, and 0.38%, respectively. However, as shown in Figures 5B,D, the survival rate changes of E. coli showed that the E. coli average survival rate of PLA-0.05, PLA-0.1, and PLA-0.2 was 33.07, 17.99, and 11.34%, respectively. With the increase in Ce-N co-doping ratio, the photocatalytic activity was enhanced, resulting in the increased production of ROS and consequently improved the antibacterial efficacy. Previous studies have indicated that photocatalytic or photodynamic antibacterial materials exhibit relatively weaker antibacterial efficacy against Gram-negative (G-) bacteria in comparison with Gram-positive (G+) bacteria (Sobotta et al., 2019; Songca and Adjei, 2022). Due to the unique structure of the cell wall in G- bacteria, it is more difficult for ROS to penetrate and exert their effects. This also explains the low efficiency of inhibition against E. coli in this section.

Figure 6 shows that PLA-N had minimal impact on the changes in optical density (OD) of the bacterial solution. However, as shown in Figure 6, the PLA-0.05, PLA-0.1, and PLA-0.2 groups exhibited growth inhibition during the logarithmic growth phase based on changes observed in OD values. In addition, the absorbance of S. aureus in PLA-0.05, PLA-0.1, and PLA-0.2 groups decreased significantly more than that of the E. coli group after irradiation.

Fluorescence images of live and dead staining are shown in Figures 7A,B. Figures 7A,B represent experimental groups under dark and light conditions, respectively. As shown in Figure 7A, there were only a few dead cells in the PLA, PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 groups under the dark condition, which proved the good biocompatibility of the membranes under the dark condition. In Figure 7B, after irradiation for 30 min, there was no significant increase in the number of dead L929 cells in the PLA group compared with dark conditions. However, in PLA-0.05, PLA-0.1, and PLA-0.2 groups, the number of dead cells after light exposure was increased compared with that under dark conditions, especially for PLA-0.1 and PLA-0.2, which was consistent with the changes in cell survival rate by CCK8 assay.

The cytotoxicity of the as-prepared PLA membranes was evaluated by a cell counting kit-8 (CCK8) viability assay against the L929 cells line as shown in Figures 7C,D. After 24-h incubation at 37°C in the dark, all PLA, PLA-N, PLA-0.05, PLA-0.1, and PLA-0.2 mats exhibited cell viabilities exceeding 90%, indicating no significant cytotoxicity toward mammalian cells. Even after 30 min of irradiation, the cell viability remained above 80%. It is noteworthy that the cell viability of the PLA-0.2 group exhibited a relatively low level, which could be attributed to the impact of photocatalytic ROS generation on cellular (Holze et al., 2018).