Integrating Porphyrinic Metal-Organic Frameworks in Nanofibrous Carrier for Photodynamic Antimicrobial Application

The rise and spread of antimicrobial resistance is creating an ever greater challenge in wound management. Nanofibrous membranes (NFMs) incorporated with antibiotics have been widely used to remedy bacterial wound infections owing to their versatile features. However, misuse of antibiotics has resulted in drug resistance, and it remains a significant challenge to achieve both high antibacterial efficiency and without causing bacterial resistance. Here, the ‘MOF-first’ strategy was adopted, the porphyrinic metal-organic frameworks nanoparticles (PCN−224 NPs) were pre-synthesized first, and then the composite antibacterial PCN−224 NPs @ poly (ε-caprolactone) (PM) NFMs were fabricated via a facile co-electrospinning technology. This strategy allows large amounts of effective MOFs to be integrated into nanofibers to effectively eliminate bacteria without bacterial resistance and to realize a relatively fast production rate. Upon visible light (630 nm) irradiation for 30 min, the PM−25 NFMs have the best 1O2 generation performance, triggering remarkable photodynamic antibacterial effects against both S. aureus, MRSA, and E. coli bacteria with survival rates of 0.13%, 1.91%, and 2.06% respectively. Considering the photodynamic antibacterial performance of the composite nanofibrous membranes functionalized by porphyrinic MOFs, this simple approach may provide a feasible way to use MOF materials and biological materials to construct wound dressing with the versatility to serve as an antibacterial strategy in order to prevent bacterial resistance.


Introduction
Antibiotic-resistant bacteria (ARB) infection has become a global crisis in wound management, which causes a delay in healing and a corresponding spike in healthcare expenses [1][2][3]. The World Health Organization (WHO) estimates that if no action is taken, the global annual cost of ARB could increase to US$100 trillion and 10 million deaths by 2050, which is far exceeding the number of cancer-caused deaths [4][5][6][7]. To overcome the arduous challenge of ARB infection, organic quaternary ammonium compounds [8][9][10][11], antimicrobial peptides [12,13], metal ions [14,15], and metal oxide nanoparticles (NPs) [16,17] have been extensively applied as alternative bactericidal agents to combining with the wound dressings. However, their existing shortcomings of certain cytotoxicity, material instability, or even increased risk of bacterial resistance in practice cannot be ignored [18][19][20]. Therefore, what is vital to wound management is the development of new therapeutics less prone to resistance.
Photodynamic therapy (PDT), an invasive approach with high spatiotemporal accuracy has become an alternative weapon against ARB [21,22]. During PDT, the photosensitizers (PS) enriched in the wounds can be activated by appropriate light, resulting optimized ex-situ. Consequently, the PCL nanofibrous membranes containing PCN−224 NPs (PM) NFMs serve as an antibacterial wound dressing that can generate 1 O2 under the 630 nm red light irradiation to attack bacteria. Such PM NFMs would be potential candidates to quickly eliminate bacteria at the wound site while preventing the development of ARB.

Synthesis of PCN−224 Nanoparticles (NPs)
The synthetic procedure to obtain PCN−224 NPs was according to the solvothermal method reported in the previously published protocol [50]. Briefly, 10 mL ZrOCl2·8H2O solution (15 mg/mL, solvent is DMF), 20 mL TCPP solution (2.5 mg/mL, DMF) and 20 mL benzoic acid solution (70 mg/mL, DMF) were dissolved uniformly under ultrasonic shaking, respectively. Then we added them to a 250 mL round bottom flask in sequence and the mixed solution was stirred at 90 °C for 5 h in an oil bath. After the reaction, the compound was allowed to cool to room temperature, the product was collected by centrifugation at 12,000 rpm for 30 min and followed by washing with DMF three times to remove unreacted substances. The final PCN−224 NPs were resuspended in fresh DMF and stored in the dark.

Synthesis of PCN−224 Nanoparticles (NPs)
The synthetic procedure to obtain PCN−224 NPs was according to the solvothermal method reported in the previously published protocol [50]. Briefly, 10 mL ZrOCl 2 ·8H 2 O solution (15 mg/mL, solvent is DMF), 20 mL TCPP solution (2.5 mg/mL, DMF) and 20 mL benzoic acid solution (70 mg/mL, DMF) were dissolved uniformly under ultrasonic shaking, respectively. Then we added them to a 250 mL round bottom flask in sequence and the mixed solution was stirred at 90 • C for 5 h in an oil bath. After the reaction, the compound was allowed to cool to room temperature, the product was collected by centrifugation at 12,000 rpm for 30 min and followed by washing with DMF three times to remove unreacted substances. The final PCN−224 NPs were resuspended in fresh DMF and stored in the dark.

Preparation of the PCL Nanofibrous Membranes Containing PCN−224 Nanoparticles (PM NFMs)
The PM NFMs were fabricated by the co-electrospinning method. Firstly, PCL was dissolved in a mixture solvent of CHCl 3 and DMF (1:1, v/v) to get a homogeneous solution with a concentration of 15 wt%. After the PCL was completely dissolved, PCN−224 NPs of different mass ratios were added, and the mixed solution was ultrasonic for 30 min and then placed on a magnetic mixer for 12 h until the PCN−224 NPs were evenly dispersed in the spinning solution. The homogeneous polymer solutions were store in a 10 mL syringe with a 22 G needle, followed by electrospinning at a voltage of 23 kV and feed rate of 0.9 mL/h. The nanofibers were collected on a collector 15 cm away from the tip of the needle. The entire electrospinning process was carried out under the temperature and relative humidity were 22 ± 2 • C and 44 ± 2%, respectively. The obtained the PCL nanofibers containing PCN−224 NPs were denoted as PM NFMs (PM−0, PM−10, PM−25 and PM−40) based on the weight ratios of PCN−224 NPs relative to PCL (0%, 10%, 25% and 40%, w/w). The PM NFMs were vacuum-dried for 12 h at 30 • C to remove the residual solvent in the fiber membrane and then protected from light.

Detection of 1 O 2 Formation
1 O 2 detection was carried out by using 1,3-diphenylisobenzofuran (DPBF) as a chemical probe [32,49,51]. Briefly, the PCN−224 NPs were incubated with 3 mL of DMF solution with a concentration of 10 µg/mL DPBF. Then the cuvette was placed under visible light of 630 nm (100 mW/cm 2 ) for 10 min, and the absorbance of the DPBF solution at 415 nm was measured every 2 min. Pure DPBF solution was used as a control sample. The test for PM NFMs is the same as PCN−224 NPs, except that the reaction medium was changed from DMF to methanol.

Photodynamic Antibacterial Assay
Gram-negative E. coli (ATCC 25922), Gram-positive S. aureus (ATCC 29213), and MRSA (ATCC 43300) were selected as representative strains to evaluate the photodynamic antibacterial ability of PM NFMs. In short, bacterial strains were cultivated overnight in Luria-Bertani (LB) medium with shaking at 37 • C. Subsequently, the concentration of the bacterial solution was diluted to 10 7 CFU/mL with PBS buffer (0.1 M, pH = 7.0). Then we drew 100 µL of bacterial solution (10 7 CFU/mL) to add to the prepared PM NFMs (1 × 1 cm 2 ), pre-cultured in the dark for 30 min, and irradiated it under visible light of 630 nm (100 mW/cm 2 ) for 30 min. The control sample was incubated in the dark for 60 min. Afterward, 0.9 mL PBS was added to each sample, and the bacteria adhered to the membranes were removed by sonication for 15 min. Then, the bacteria supernatant was 10-fold gradient diluted with PBS, and 100 µL of each dilution gradient was inoculated on LB agar plates. The treated agar plate was placed in a 37 • C constant temperature and humidity incubator for 18-24 h. We counted the number of colonies on the agar plate to evaluate its photodynamic antibacterial performance. This was done for each sample three times.

Cytotoxicity Assay
A standard CCK-8 assay was selected to evaluate the cytotoxicity of PM NFMs. First, L929 cells were cultured in the complete medium containing 90% DMEM, 10% fetal bovine serum, and 1% penicillin-streptomycin solution. Subsequently, the cells were inoculated on 14 mm PM NFMs at an implantation density of 2 × 10 4 cells per well, the wells without samples were used as controls, and the well plates were placed in the incubator after inoculation was completed. After 24 h incubation, the medium was aspirated from each well and the cells were washed three times with PBS to remove unadhered free cells. Finally, 500 µL of CCK-8 working solution was added to each well under dark conditions, and the absorbance of the supernatant at 450 nm was measured after 3 h of incubation at 37 • C incubators. This was done for each sample three times.

Characteristics of PCN−224 NPs
In our 'MOF-first' strategy, MOFs should be pre-synthesized firstly and then mixed with the polymer solution for fiber processing. As shown in Figure 1a, PCN−224 NPs consist of the six-connected Zr6 clusters and the four carboxyl porphyrin ligand TCPPs through coordination; each Zr6 cluster bridges six TCPP ligands forming the spacious three-dimensional nanochannel framework [52]. The morphological analysis of the pre-synthesized PCN−224 NPs were characterized by SEM and TEM. As shown in Figure 1b,c, PCN−224 NPs present a typical spherical appearance with an average diameter of 79.02 ± 0.58 nm ( Figure S1) and good dispersity in the solution. To analyze whether the porphyrins were introduced into the frameworks, the TEM elemental mapping of PCN−224 NPs were examined (Figures 1d and S2). The appearance and even distribution of Zr, O, and N elements, indicating that the existence of porphyrin throughout the whole framework preliminarily [53]. Furthermore, the calculated ratio of Zr and porphyrin in the MOF is about 3.18, which is comparable to the ratio in the theoretical synthetic framework (by the structure of PCN−224, the theoretical ratio of Zr is 4 of that of TCPP) [53][54][55]. Hereafter, N 2 adsorption-desorption isotherm measurement was used to reveal the porous structure of the PCN−224 NPs. As demonstrated in Figure 1e, the PCN−224 NPs had a BET surface area of 837.66 m 2 /g, and the pore size distribution is mainly concentrated at 1.43 nm (Figure 1f). It can be seen that the introduction of porphyrin did not sacrifice the porous features of MOFs, which would be conducive to the efficient diffusion of 1 O 2 [56,57].   Attempting to identify the crystallographic structure of pre-synthesized PCN−224 NPs, XRD analysis was performed. As shown in Figure 2a, the characteristic diffraction peaks of the synthesized PCN−224 NPs were in excellent agreement with the simulation pattern at the angles of 3 • to 20 • , proving the crystalline phase purity. The diffraction peaks of PCN−224 NPs at 2θ = 4.46 • , 6.46 • , 7.78 • , 8.98 • , and 11.20 • which represent the (002), (022), (222), (004) and (224) crystal planes, respectively [58], also indicating that the porphyrinrelated inhibition effect on crystal formation was not displayed. In the FTIR spectra (Figure 2b), the peak at 1700 cm −1 corresponds to C=O stretching band. The strong vibration assigned to C=O was found in the peak of 1656 cm −1 , which was owing to the coordination with metal ions [49]. The C=C bond stretching vibration peak of benzene and pyrrole ring appeared at 1552-1601 cm −1 [59]. Furthermore, it is essential to observe a peak at 658 cm −1 which was owing to the Zr-OH bond vibration [49]. The XPS spectrum of Zr 3d ( Figure S3) shows that compared with the standard spectrum data (182.4 ev), the Zr 3d 5/2 peak shifts 0.2 eV toward the lower binding energy. This shift indicates the charge redistribution within Zr6 nodes in PCN−224 NPs, suggesting the successful coordination of the carboxyl group of the porphyrin to unsaturated sites of Zr6 [60]. The successful synthesis of PCN−224 NPs was demonstrated by the above results.
of the carboxyl group of the porphyrin to unsaturated sites of Zr6 [60]. The successful synthesis of PCN−224 NPs was demonstrated by the above results.

Studying Activity of PCN−224 NPs
With an indication of the porphyrin introduction, the question remained as to whether the porphyrin retained its photodynamic functionality. We thus carried out the UV-Vis analysis to characterization PCN−224 NPs. As illustrated in Figure 2c, PCN−224 NPs demonstrated the main absorption peak (Soret band) at 420 nm and four weak absorption peaks (Q bands) in a range of 500-700 nm, similar to the TCPP spectrum. The Q bands of TCPP were indicated that the Zr ions were not coordinated to the center of the porphyrin ligands [61]. In contrast, the Soret band of PCN−224 is significantly weaker and demonstrates a slight redshift, which may be due to ligand-to-metal charge transfer caused by the strong coordination of the TCPP linkers to the Zr-O clusters in the PCN−224 framework [62,63]. The photoluminescence (PL) spectra show two strong emission bands of PCN−224 NPs at about 658 nm and 721 nm under an excitation wavelength of 430 nm, which are typical emission peaks for the transition of porphyrins from S 1 →S 0 states, corresponding to Q (0-0) and Q (0-1) transitions, respectively ( Figure S4) [64,65]. This absorption and PL spectra pattern are the well-known characteristic of the monomeric porphyrin molecules [66]. This demonstrates the structural integrity of the porphyrin ligand in the PCN−224 NPs framework, while showing as well that red fluorescence can be used as an imaging label for PCN−224 NPs [49]. In addition, the PL spectrum underwent a similar change to the UV-Vis. This may be caused by the coordination of the TCPP ligand to the Zr6 cluster changing the molecular planarity in the porphyrin macrocycle and affecting the distribution of the porphyrin electron density [63,67].
1 O 2 is the most destructive bacterial substance in ROS. To further investigate the PDT activities of PCN−224 NPs, the generation capabilities of 1 O 2 from PCN−224 NPs under a visible light lamb were measured by using the DPBF probe. As depicted in Figures 2d and S5, the absorbance at 415 nm of pure DPBF solution displayed a slight decrease after 10 min of irradiation. This phenomenon is due to the self-decomposition of DPBF since it is extremely sensitive to external light [39]. By contrast, the absorbance of the DPBF solution with the addition of PCN−224 NPs has significant decay trends in the same conditions, demonstrating their 1 O 2 generation capability. These results prove the effectiveness of the one-step synthesis route for the preparation of PCN−224 NPs, which have important potential as a functional photosensitizer carrier and PDT agent for the treatment of bacterial infections.

Nanofiber Fabrication with Preformed PCN−224
Co-electrospinning offers a facile route for loading MOFs in polymeric fibers. Low MOFs loadings can facilitate the maintenance of the structural integrity of the NFMs morphology and three-dimensional porous structure, however, there have been examples of NFMs, that exhibit improved antibacterial and gas adsorption performance when the MOFs content in the NFMs was increased [39,68]. Meanwhile, the chemical differences between MOFs crystals and polymers can manifest as defects (such as stability, thermal stability, etc.) in the composite NFMs [69]. Accordingly, we use different concentrations of PCN−224 NPs to fabricate NFMs. The surface morphology of the nanofibers after assembly with preformed PCN−224 NPs was observed in Figure 3. As shown in Figure 3a-d, the surface of neat PCL nanofibers (PM−0) was smooth, whereas after PCN−224 NPs were introduced, the surface morphology of nanofibers changed obviously. Compared with pristine PM−0 nanofibers, only a few PCN−224 NPs were sporadically distributed on the surface of PM−10 nanofibers, which due to the lowest loading amount of PCN−224 NPs and the NPs are mainly distributed in the fiber matrix. It is noteworthy that as the concentration of the PCN−224 NPs increases, the NPs on the surface of the nanofibers were sharply increased, and even formed agglomerates, resulting in the PM−25 and PM−40 nanofibers showing a significantly rougher surface morphology. Meanwhile, it is observed that a dramatic increase in the nanofiber diameter from 0.25 to 1.85 µm with the gradual increment in the dosage of PCN−224 NPs (Figure 3e-h). This phenomenon may be ascribed to the enhanced viscosity of the spinning solution with the incorporation of PCN−224 NPs, which prolongs the relaxation time and restricts the motion of the polymer chains [32]. be ascribed to the enhanced viscosity of the spinning solution with the incorporation of PCN−224 NPs, which prolongs the relaxation time and restricts the motion of the polymer chains [32].  Figure 4e, it can be found that the asymmetric and symmetric vibrations of CH2 at 2942 cm −1 and 2862 cm −1 were ascribed to PM−0 and the C=C bond stretching vibration appeared at 1552-1605 cm −1 , which was attributed to the presence of PCN−224 NPs [70]. The FTIR results indicating that the PCN−224 NPs were successfully introduced into PCL nanofibers [58,71]. The DSC results ( Figure S6) demonstrate the possible coordination interaction between Zr metal nodes and PCL carbonyl groups under the introduction of PCN−224. [72,73]. Nevertheless, as a PDT wound dressing, the stable integration of PS NPs and nanofibers is the most staple requirement. For this reason, we immersed the PM NFMs in HEPES buffer (50 mM, pH = 7.4) solution and tested the UV-visible spectrum of the soaking solution to explore the stability of the PM NFMs ( Figure S7). The results showed that the curves of PM NFMs are completely consistent, and no characteristic peaks of PCN−224 NPs were detected, which indicates that no PCN−224 NPs were shed from the fibers and they are stably and firmly combined with PCL nanofibers. In addition, the results ( Figure S8) of XRD show that the PM−0 NFMs have two diffraction peaks at 21.5° and 23.8°, which are characteristic peaks of neat PCL, corresponding to the (110) and (200) planes, respectively [74]. The PM−25 NFMs retain the characteristic diffraction peaks of neat PCL, while the characteristic diffraction peaks of PCN−224 appear. This indicates that the structure of PCN−224 NPs in the composite membrane has been well preserved. Comparing the FTIR spectra of PM NFMs in Figure 4e, it can be found that the asymmetric and symmetric vibrations of CH 2 at 2942 cm −1 and 2862 cm −1 were ascribed to PM−0 and the C=C bond stretching vibration appeared at 1552-1605 cm −1 , which was attributed to the presence of PCN−224 NPs [70]. The FTIR results indicating that the PCN−224 NPs were successfully introduced into PCL nanofibers [58,71]. The DSC results ( Figure S6) demonstrate the possible coordination interaction between Zr metal nodes and PCL carbonyl groups under the introduction of PCN−224 [72,73]. Nevertheless, as a PDT wound dressing, the stable integration of PS NPs and nanofibers is the most staple requirement. For this reason, we immersed the PM NFMs in HEPES buffer (50 mM, pH = 7.4) solution and tested the UV-visible spectrum of the soaking solution to explore the stability of the PM NFMs ( Figure S7). The results showed that the curves of PM NFMs are completely consistent, and no characteristic peaks of PCN−224 NPs were detected, which indicates that no PCN−224 NPs were shed from the fibers and they are stably and firmly combined with PCL nanofibers. In addition, the results ( Figure S8) of XRD show that the PM−0 NFMs have two diffraction peaks at 21.5 • and 23.8 • , which are characteristic peaks of neat PCL, corresponding to the (110) and (200) planes, respectively [74]. The PM−25 NFMs retain the characteristic diffraction peaks of neat PCL, while the characteristic diffraction peaks of PCN−224 appear. This indicates that the structure of PCN−224 NPs in the composite membrane has been well preserved.
Accurate knowledge of the PCN−224 NPs loading during electrospinning is important because excess PCN−224 NPs in the polymeric solution can result in needle clogging and instability of the jets. Therefore, the TGA test was used to determine the actual loading of PCN−224 NPs in NFMs (Figure 4f). It can be clearly observed that the residual mass fraction of NFMs at 800 • C shows an increasing trend with the increase of PCN−224 NPs content. Subsequently, the calculation was performed according to previous research [75,76], and it was assumed that the only substance present at 800 • C is the secondary building unit (ZrO 2 ) derived from PCN−224 NPs; the effective loading rate of PCN−224 NPs in the NFMs was deduced by a comparison of the residual mass fraction of ZrO 2 of these NFMs. The residual mass fraction of ZrO 2 in PM−10 NFMs was about 4.48 wt%. Using this as a standard, the theoretical residual mass fractions of PM−25 and PM−40 NFMs were 10.95 wt% and 17.52 wt%, respectively. However, the test results showed that the actual residual mass fraction of 10.00 wt% for PM−25 NFMs and 13.32 wt% for PM−40 NFMs. It can be seen that there exist certain differences in theoretical and practical residuals, especially in PM−40 NFMs. This indicates that the actual loading efficiency of PCN−224 NPs in the PM−40 NFMs is low. This phenomenon was to be anticipated because excessive PCN−224 NPs will form agglomerates that are unevenly dispersed in the polymeric solution, and clog the needles during the electrospinning process, resulting in the difficult formation and poor consistent of fibers, which can also be observed in the above SEM images. Furthermore, the DTG curves revealed that although the introduction of PCN−224 NPs affects the crystal structure of the PCL polymers, the thermal degradation temperature of NFMs decreased slightly from 450 • C to 390 • C with the increase in loading amount, but its thermostability still met the requirements of practical applications.
In addition, the tensile strengths of the PM NFMs were measured to evaluate the influence of the PCN−224 NPs on the mechanical properties of PM NFMs. As shown in Figure 4g, the neat PM−0 NFM showed the highest mechanical breaking elongation of 245.45% with breaking strength at 3.72 MPa. Slightly increased breaking strength (3.86 MPa) and decreased breaking elongation (240.55%) could be observed in PM−10 NFMs. This phenomenon may be due to the coordination interaction between the carbonyl group of PCL and the Zr metal nodes of PCN−224, and the more uniform distribution of PCN−224 NPs without forming obvious weak defects, thus improving the tensile strength of the membranes to a certain extent [72,77]. After more PCN−224 NPs were introduced, the breaking strength and breaking elongation of PM−25 and PM−40 NFMs were decreased to a large extent. This is attributed to the agglomerate of PCN−224 NPs in the membranes to form stress concentration centers, which increases the weak defects. Moreover, due to the increase in PCN−224 NPs concentration, the spinnability of the nanofiber is reduced, resulting in a decrease in the uniformity of the nanofiber [77,78].

1 O 2 Generation of PM NFMs
As a proof of concept, we also demonstrated the 1 O 2 generation ability of PM NFMs. The mechanism of PM NFMs in the generation of 1 O 2 is summarized in Figure 5a. Under the excitation of the laser with appropriate wavelength, PCN−224 NPs, as a new PS, can react with 3 O 2 through energy transfer to generate cytotoxic 1 O 2 , which can kill bacteria by attacking the bacterial biomolecules [56]. The results of employing the sensing probe DPBF to investigate the capacities of PM NFMs to generate 1

The Morphologies and Distribution of PCN−224 NPs in PM NFMs
To further clarify the morphologies and distributions of PCN−224 NPs in the nanofibers, the CLSM images of each NFMs were measured. As shown in Figure 6a

The Morphologies and Distribution of PCN−224 NPs in PM NFMs
To further clarify the morphologies and distributions of PCN−224 NPs in the nanofibers, the CLSM images of each NFMs were measured. As shown in Figure 6a This type of morphology is typical, such that the resultant NFMs are non-uniform with unpredictable properties, which is consistent with the results of the 1 O 2 generation performance. It can be observed from Figure 6b that the red fluorescence signal of NFMs was enhanced with the content of PCN−224 NPs increases. These results powerfully confirmed that the distribution and amount of PCN−224 NPs in the membrane affect its ability to generate 1 O 2 .

In Vitro Antibacterial Activity Assay
Gram-positive S. aureus, Gram-negative E. coli, and drug-resistant bacteria MRSA were selected as representative strains, and the photodynamic antibacterial performance of the PM NFMs were evaluated by the method of counting live bacteria on the plate. Initially, the antibacterial ability of PM NFMs was visually and qualitatively evaluated through monitoring the photos of the agar plate coated with the residual bacteria after PM NFMs treatment (Figure 7a,c), respectively. In the dark, all membranes groups showed densely distributed colonies on the agar plate, indicating that all the membranes had no antibacterial activity in the darkest condition. In contrast, the membranes introduced with PCN−224 NPs under light irradiation exhibited efficient photodynamic antibacterial effects on S. aureus and E. coli, as indicated by the sparsely distributed bacterial colonies. Furthermore, the survival rate of bacteria was calculated to quantitatively evaluate the antibacterial activity of the PM NFMs (Figure 7b,d). Compared with PM−0 NFMs, the PM−10 NFMs exhibited slightly antibacterial activity against S. aureus and E. coli within 30 min light irradiation due to the lowest loading amount of PCN−224 NPs. The corresponding survival rates of 21.04% for S. aureus and 52.51% for E. coli. An improvement in antibacterial activity was noted as the increased concentration of PCN−224 NPs contained in the membranes. In detail, the PM−25 NFMs exhibited excellent potent antibacterial properties with survival rates of 0.13% for S. aureus and 2.06% for E. coli upon 30 min irradiation, respectively. For the PM−40 NFMs, the survival rates of S. aureus and E. coli are 6.48% and 11.04% under 30 min irradiation, respectively. It could be clearly observed that the antibacterial ability of PM−25 NFMs was superior to the PM−40 NFMs, which was in accordance with the previous result of 1 O2 generation. These results confirmed that the prepared membranes containing PCN−224 NPs have significant antibacterial activities, and also demonstrated that the antibacterial activity is strongly associated with the loading dose and distribution of PCN−224 NPs. The bacteria were attached to the surface of the composite membrane, and the nanofiber morphology did not change significantly under the dark/light treatment, which indicated that the bacteria would not affect the nanofiber structure ( Figure S10). It is also worth mentioning that the different sensitivities to 1 O2 between Gram-positive S. aureus and Gram-negative E. coli. The Gram-negative bac-

In Vitro Antibacterial Activity Assay
Gram-positive S. aureus, Gram-negative E. coli, and drug-resistant bacteria MRSA were selected as representative strains, and the photodynamic antibacterial performance of the PM NFMs were evaluated by the method of counting live bacteria on the plate. Initially, the antibacterial ability of PM NFMs was visually and qualitatively evaluated through monitoring the photos of the agar plate coated with the residual bacteria after PM NFMs treatment (Figure 7a,c), respectively. In the dark, all membranes groups showed densely distributed colonies on the agar plate, indicating that all the membranes had no antibacterial activity in the darkest condition. In contrast, the membranes introduced with PCN−224 NPs under light irradiation exhibited efficient photodynamic antibacterial effects on S. aureus and E. coli, as indicated by the sparsely distributed bacterial colonies. Furthermore, the survival rate of bacteria was calculated to quantitatively evaluate the antibacterial activity of the PM NFMs (Figure 7b,d). Compared with PM−0 NFMs, the PM−10 NFMs exhibited slightly antibacterial activity against S. aureus and E. coli within 30 min light irradiation due to the lowest loading amount of PCN−224 NPs. The corresponding survival rates of 21.04% for S. aureus and 52.51% for E. coli. An improvement in antibacterial activity was noted as the increased concentration of PCN−224 NPs contained in the membranes. In detail, the PM−25 NFMs exhibited excellent potent antibacterial properties with survival rates of 0.13% for S. aureus and 2.06% for E. coli upon 30 min irradiation, respectively. For the PM−40 NFMs, the survival rates of S. aureus and E. coli are 6.48% and 11.04% under 30 min irradiation, respectively. It could be clearly observed that the antibacterial ability of PM−25 NFMs was superior to the PM−40 NFMs, which was in accordance with the previous result of 1 O 2 generation. These results confirmed that the prepared membranes containing PCN−224 NPs have significant antibacterial activities, and also demonstrated that the antibacterial activity is strongly associated with the loading dose and distribution of PCN−224 NPs. The bacteria were attached to the surface of the composite membrane, and the nanofiber morphology did not change significantly under the dark/light treatment, which indicated that the bacteria would not affect the nanofiber structure ( Figure S10). It is also worth mentioning that the different sensitivities to 1 O 2 between Gram-positive S. aureus and Gram-negative E. coli. The Gram-negative bacteria showed more tolerance to 1 O 2 than Gram-positive bacteria, which is ascribed to the different structures of bacterial cell walls. The Gram-negative bacteria contain an additional layer of lipopolysaccharide on the outside of the peptidoglycan layer, which has a high degree of impermeable, making it difficult for 1 O 2 to enter the inside of the bacteria, and cannot destroy bacterial biomolecules such as proteins, nucleic acid and lipids [58].
MRSA is a typical example, it has become one of the most prevalent pathogens in clinic wound infections [79][80][81]. The local skin infection caused by it is almost resistant to all conventional antibiotics, greatly increases the difficulty of clinical treatment, and seriously threatens public health [82]. Hence, MRSA was selected as the representative strain of drug-resistant bacteria to evaluate the performance of PM NFMs in eliminating drugresistant bacteria ( Figure S11). The results show that the PM−25 NFMs exhibited excellent antibacterial properties with survival rates of 1.91% for MRSA upon 30 min irradiation. The antibacterial test results show that the photodynamic composite membranes prepared by introducing PCN−224 NPs can effectively eliminate drug-resistant bacteria and common bacteria that are widely present in the wound, and has excellent antibacterial effects.
Polymers 2021, 13, x FOR PEER REVIEW 13 of 18 different structures of bacterial cell walls. The Gram-negative bacteria contain an additional layer of lipopolysaccharide on the outside of the peptidoglycan layer, which has a high degree of impermeable, making it difficult for 1 O2 to enter the inside of the bacteria, and cannot destroy bacterial biomolecules such as proteins, nucleic acid and lipids [58]. MRSA is a typical example, it has become one of the most prevalent pathogens in clinic wound infections [79][80][81]. The local skin infection caused by it is almost resistant to all conventional antibiotics, greatly increases the difficulty of clinical treatment, and seriously threatens public health [82]. Hence, MRSA was selected as the representative strain of drug-resistant bacteria to evaluate the performance of PM NFMs in eliminating drugresistant bacteria ( Figure S11). The results show that the PM−25 NFMs exhibited excellent antibacterial properties with survival rates of 1.91% for MRSA upon 30 min irradiation. The antibacterial test results show that the photodynamic composite membranes prepared by introducing PCN−224 NPs can effectively eliminate drug-resistant bacteria and common bacteria that are widely present in the wound, and has excellent antibacterial effects.

Cytotoxicity Assay
Last but not least, we used the CCK-8 method to measure the viability of L929 cells co-cultured with PM−0 and PM−25 NFMs for 24 h to evaluate the cytotoxicity of the composite membrane. As shown in Figure 8, the viability of the cells on the PM−0 NFMs reached 123%, indicating that the neat PCL membrane has excellent biocompatibility, which is consistent with the results of previous studies. The cells co-cultured with PM−25 NFMs also maintained high viability (85%), which indicates that the nanocomposite membrane after the introduction of PCN−224 NPs has good biocompatibility, which is beneficial for its application in anti-infection and wound healing.

Cytotoxicity Assay
Last but not least, we used the CCK-8 method to measure the viability of L929 cells co-cultured with PM−0 and PM−25 NFMs for 24 h to evaluate the cytotoxicity of the composite membrane. As shown in Figure 8, the viability of the cells on the PM−0 NFMs reached 123%, indicating that the neat PCL membrane has excellent biocompatibility, which is consistent with the results of previous studies. The cells co-cultured with PM−25 NFMs also maintained high viability (85%), which indicates that the nanocomposite membrane after the introduction of PCN−224 NPs has good biocompatibility, which is beneficial for its application in anti-infection and wound healing.

Conclusions
In conclusion, we adopted the 'MOF-first' strategy to pre-synthesize porphyrinic MOFs nanoparticles (PCN−224 NPs) firstly, then mixed with the polymer solution to fiber processing via one-step co-electrospinning technology, and successfully developed PM NFMs with PDT antibacterial properties. The structure of PCN−224 NPs can be well preserved after integration with PCL, and a large number of firm loads can be achieved on the nanofibers. Both PCN−224 NPs and PM NFMs showed excellent ability of 1 O2 generation. The PM NFMs could increase the 1 O2 generation and exhibit excellent PDT antibacterial against S. aureus, E. coli, and MRSA in vitro based on the increased loading of PCN−224 NPs. Importantly, the PM−25 NFMs had the best PDT antibacterial performance with survival rates of 0.13% for S. aureus, 2.06% for E. coli, and 1.91% for MRSA upon 30 min irradiation, due to a homogeneous and stable loading of PCN−224 NPs in nanofibers. The cytotoxicity assay verified that the PM NFMs possessed good biocompatibility. Taken together, these NFMs functionalized by porphyrinic MOFs could be regarded as a promising wound dressing, and could be widely applied in the field of wound antibacterial infection.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Figure S1: The size distribution of the PCN−224 NPs, Figure S2: The transmission electron microscopy (TEM) elemental mapping results of PCN−224 NPs, Figure S3: XPS spectra of PNC-224 and Zr 3d, Figure  S4: The PL spectra of PCN−224 NPs, Figure S5

Conclusions
In conclusion, we adopted the 'MOF-first' strategy to pre-synthesize porphyrinic MOFs nanoparticles (PCN−224 NPs) firstly, then mixed with the polymer solution to fiber processing via one-step co-electrospinning technology, and successfully developed PM NFMs with PDT antibacterial properties. The structure of PCN−224 NPs can be well preserved after integration with PCL, and a large number of firm loads can be achieved on the nanofibers. Both PCN−224 NPs and PM NFMs showed excellent ability of 1 O 2 generation. The PM NFMs could increase the 1 O 2 generation and exhibit excellent PDT antibacterial against S. aureus, E. coli, and MRSA in vitro based on the increased loading of PCN−224 NPs. Importantly, the PM−25 NFMs had the best PDT antibacterial performance with survival rates of 0.13% for S. aureus, 2.06% for E. coli, and 1.91% for MRSA upon 30 min irradiation, due to a homogeneous and stable loading of PCN−224 NPs in nanofibers. The cytotoxicity assay verified that the PM NFMs possessed good biocompatibility. Taken together, these NFMs functionalized by porphyrinic MOFs could be regarded as a promising wound dressing, and could be widely applied in the field of wound antibacterial infection.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/ 10.3390/polym13223942/s1, Figure S1: The size distribution of the PCN−224 NPs, Figure S2: The transmission electron microscopy (TEM) elemental mapping results of PCN−224 NPs, Figure S3: XPS spectra of PNC-224 and Zr 3d, Figure S4: The PL spectra of PCN−224 NPs, Figure S5: UV-visible spectrum of DPBF under visible illumination (630 nm, 100 mW/cm 2 ) with/without PCN−224 NPs, Figure S6. DSC scans of PM−0 and PM−25 NFMs. Left) heating and Right) cooling, Figure S7: The UV-visible spectrum of PM NFMs after immersing in HEPES buffer for 12 h, Figure S8: XRD patterns of PM−0 and PM−25 NFMs, Figure S9: UV-visible spectrum of DPBF under visible illumination (630 nm, 100 mW/cm 2 ), Figure S10. The SEM images of bacteria and bacteria treated fibers, Figure S11: Relative bacterial survival rate and photographs of residual colonies of MRSA under various membranes treatments in the dark/light.