Stepwise antibacterial strategy for orthopedic implants using bacteriophages on electrospun cefiderocol/PCl/Gt nanofibers over PEO-coated Mg3ZnCa

Implant-associated infections represent one of the most severe complications in orthopedics, often leading to implant failure, osteomyelitis, or septic arthritis. Despite numerous preventive measures, periprosthetic infections remain a major clinical challenge. The pathogenesis typically begins with bacterial adhesion and colonization on implant surfaces, resulting in biofilm formation. Biofilms serve as a protective shield against antibiotics and host immune responses. Over the last decade, preventive measures have helped to reduce the incidence of orthopedic implant infections¹. Still, the growing number of arthroplasties in elderly patients and the complexity of surgical reconstructions continue to increase the risk of infection². It is reported that yearly, about 1 million hip and knee arthroplasties are operated on in the U.S., a figure projected to be 4 times more within 10–20 years as the population ages³. Despite a seemingly low 1–2% incidence, periprosthetic joint infection is now the leading cause of revision surgery, carrying major physical, emotional, and economic burdens, and contributing to increased mortality. Consequently, recent strategies emphasize localized infection control at the implant–tissue interface. One promising approach is the incorporation of antimicrobial coatings on implants, enabling targeted drug release directly into peri-implant tissues. Preclinical and clinical evidence support the efficacy of such coatings–using agents like silver, gentamicin, or iodine–in significantly reducing the incidence of postoperative periprosthetic infections. Thus, antimicrobial surface modifications may represent a pivotal step in both prevention and treatment of implant-related infections^4,5,6,7.

Implants made from magnesium alloys offer significant advantages in biomedical applications due to their biodegradability, biocompatibility, and osteoconductivity. Unlike traditional metal implants, magnesium gradually dissolves in the body, eliminating the need for surgical removal and reducing long-term complications^8,9. Its stiffness closely resembles natural bone, thus minimizing stress shielding and promoting better integration with surrounding tissue⁹. Additionally, magnesium ions stimulate bone regeneration and create an antibacterial environment, lowering the risk of infections^10,11. However, its rapid corrosion in physiological environments poses a significant limitation, as uncontrolled degradation undermines implant stability before sufficient healing has been achieved^12,13. To address this issue, surface modification techniques, such as plasma electrolytic oxidation (PEO), have been extensively employed to enhance the corrosion resistance of Mg alloys¹⁴. This method is very similar to anodizing; the difference is that the high-voltage power supply creates glowing sparks and turns the surface material into a ceramic layer^{14,15,16,17,18,19,20}. Moreover, in vivo studies have highlighted the potential of PEO for bone tissue engineering²¹. However, while PEO provides advantageous corrosion protection, it lacks antibacterial properties, making implants susceptible to bacterial colonization and biofilm formation²². Incorporating antimicrobial agents onto the PEO surface may impart antibacterial activity and thereby address this limitation.

Electrospinning has emerged as a versatile method for creating nanofiber coatings that incorporate and gradually release therapeutic agents, including antibiotics²³. This technique uses electric forces to draw charged polymer solutions into fibers, typically ranging from nanometers to a few micrometers in diameter²⁴. Electrospun nanofibers are widely applied in tissue engineering, drug delivery, and filtration membranes^24,25.

Many studies have combined polycaprolactone (PCL) and gelatin (Gt) in electrospun nanofibers, incorporating antibiotics for antibacterial properties^26,27. Sezer et al.²⁸ explored PCL/Gt nanofibers with gentamicin for bone tissue engineering. They developed biodegradable, osteoconductive scaffolds with multifunctional capabilities, including antibiotic delivery. Bakhsheshi-Rad et al.²⁹ loaded electrospun nanofibers with ciprofloxacin and deposited the fibers on the surface Mg-1Ca alloy. Although the electrospun nanofibers exhibited the desired drug release and cell culture properties, the porosity of the electrospun layer exposed the underlying magnesium alloy to the corrosive ambient conditions. The weak corrosion resistance of the substrate combined with gas production during the corrosion process led to a reduction in the number of cultured cells.

While gentamicin and ciprofloxacin are well-known antibiotics, cefiderocol, used in the current study, is a novel antibiotic with minimal resistance concerns^30,31. It exhibits strong bactericidal properties against Gram-negative bacteria and would be advantageous to develop a therapeutic strategy. E. coli and Pseudomonas aeruginosa (P. aeruginosa) are the most frequent causes of orthopedic implant infections among Gram-negative bacteria (20–30%) and are particularly important because of their severity and the high risk of treatment failure^32,33. Periprosthetic joint infections caused by Gram-negative bacteria, such as E. coli, are being reported with increasing frequency. Eradication of these infections may require up to one year of follow-up³⁴.

Although antibiotics are the most well-known medicines for combating infections, issues such as antibiotic resistance, multi-drug resistant (MDR) infections, and their adverse effects on organs have driven scientific research toward exploring bacteriophages (phages) as well. Phages are viruses that target and kill specific bacteria. They offer key advantages, like biofilm penetration and synergy with antibiotics, making them promising options^35,36,37,38. Phages have diverse shapes and sizes, but the most well-known and studied ones, such as Myoviridae, Siphoviridae, and Podoviridae, generally have an icosahedral (20-sided) head and a tail. The head, also called the capsid, is typically 50 to 200 nm in diameter. The size of the tails can vary significantly depending on the phage family, between 10 and 200 nm in length, and 5 to 20 nm in thickness^39,40. Imaging of immobilized phages is an open topic in this field, mainly because of their small size^35,41,42, which is one of the subjects we have tackled in this study.

Fanaei Pirlar et al.⁴³ assessed the effectiveness of cefiderocol combined with bacteriophages against E. coli, Klebsiella pneumoniae, P. aeruginosa, and Acinetobacter baumannii. They demonstrated that cefiderocol and phages exhibit a synergistic effect in inhibiting bacterial growth, meaning that their combined use is more effective than using either agent alone. The current study puts the knowledge of the synergic effect of cefiderocol and phages⁴³ in a novel applicational method to modify the antibacterial implants. E. coli ATCC 25922 (a Gram-negative bacterium) was selected as a model organism to demonstrate the feasibility of our novel cefiderocol-phage coating method. The use of an ATCC reference strain facilitates reproducibility and comparability across different laboratories.

This study introduces a multifunctional, stepwise method for orthopedic implants by integrating PEO-coated Mg3ZnCa (ZX31) with cefiderocol-loaded electrospun nanofibers and phages immobilized on the nanofibers. Applying a PEO coating before nanofiber deposition on ZX31 effectively mitigated corrosion of the magnesium alloy. Additionally, an antibiotic (cefiderocol) is incorporated into the electrospinning solution to provide controlled antibacterial functionality, beside immobilized phages on the nanofibers, actively combating postsurgical infections. The comprehensive characterization and testing of the coatings, including corrosion performance, drug release profiles, and antibacterial efficacy, highlight the potential of this novel approach to significantly reduce the risk of implant-related infections. This strategy offers a promising solution to one of the most critical challenges in orthopedic surgery, paving the way for safer and more effective implant technologies.

Step 1: PEO coating; characterization

PEO results in a typical rough and porous surface layer on the ZX31 alloy⁸ (Fig. 1a). The cracks (marked by arrows) are due to shrinkage during solidification. The cross-section through the coated sample (Fig. 1b) highlights that the PEO layer consists of two parts: an inner, dense layer (marked in orange) and an outer one (marked in purple), containing the pores. Some of these pores seem to be connected by channels, and they reach deep into the outer layer. The dense inner layer is important for corrosion protection of the substrate⁴⁴. Beneath the PEO coating, we see the alloy with its typical eutectic microstructure, consisting of a-Mg grains, surrounded by Ca2Mg6Zn3 precipitations at the grain boundaries⁴⁵. Figure 1c shows current versus time during the PEO process. As explained by many studies^13,45,46, after the current‒time diagram passes the sharp peak, the PEO process remains in an almost stable range of current. Thus, here in the first 10 s, the surface reached the stable plasma level, and the coating was completed during the process time.

Figure 1d depicts the potentiodynamic polarization curves of ZX31 and ZX31PEO in PBS, highlighting the corrosion protection effect of the PEO coating in a corrosive environment. The data derived from the polarization curves are summarized in Table 1. The corrosion current density (I_corr) for the ZX31PEO samples was approximately 15.30 μA/cm², whereas for the uncoated samples, it was 31.23 μA/cm². This disparity demonstrates the positive effect of the PEO coating on the corrosion behavior of the samples. The I_corr of the ZX31PEO sample is reduced to 50% compared to the uncoated sample, and, concurrently, the corrosion resistance of the ZX31 sample is improved up to twofold due to the PEO coating. Accordingly, the corrosion potential (E_corr), shifted toward a more positive value from −1.66 V for ZX31, to −1.42 V for ZX31PEO. A more positive E_corr means that the material is less prone to oxidation and dissolution in a corrosive environment. The corrosion rate corroborates the findings. It is reduced by nearly 51% after PEO coating, indicating improved durability of the material.

Step 2: Electrospun nanofibers on the PEO coating; characterization and application

The electrospun nanofibers cover the PEO coating as the second step. Figure 2a, Figure 2b, c present typical SEM micrographs of the electrospun nanofibers of CFD6, CFD12, and CFD24. The synthesized nanofibers are straight and randomly placed over and next to each other. The thickness of the fibers with an average diameter of 53.05 nm and a standard deviation of 18.45 nm was evaluated as CFD 6 to be the least uniform sample. The CFD 12 shows a better homogeneity in fiber diameter along the fibers (average diameters: 99.31 nm, standard deviations: 64.52). Finally, for CFD24, it can be observed that there are two classes of very thin and very thick nanofibers. CFD 24 fibers are mainly thicker (average diameters: 110.10 nm, standard deviations: 53.15) than those of the CFD6 and CFD12 nanofibers due to the higher percentage of CFD. The important point is that the surface of all nanofibers is smooth. The nanofibers of CFD 24 are the smoothest. Then CFD 12 and CFD 6 are less smooth compared to CFD 24, respectively.

Figure 2d exhibits the FTIR spectra of the CFD12 nanofibers as the representative nanofiber. The FTIR diagram shows bonds related to PCL and Gt groups at 2932 cm⁻¹ (asymmetric CH₂ stretching), 2854 cm⁻¹ (symmetric CH₂ stretching) often linked to aliphatic groups, such as the methylene (CH₂) groups in cefiderocol, 1729 cm⁻¹ (carbonyl stretching), 1635 cm⁻¹ (C = O stretching), 1524 cm⁻¹ amide II band, which involves N–H bending and C–N stretching, 1292 cm⁻¹ (C-O and C-C stretching), 1242 cm⁻¹ (asymmetric C-O-C stretching), 1170 cm⁻¹ (symmetric C-O-C stretching), and 724 cm⁻¹ (N-H out-of-plane wagging)⁴⁶. The peak at 1462 cm⁻¹ is often associated with CH₂ bending vibrations⁴⁷. In PCL and gelatin, the peak at 1363 cm⁻¹ typically corresponds to CH₃ bending vibrations, which are often seen as symmetric bending or deformation of methyl groups. It can also be related to the bending vibrations of -OH groups. For cefiderocol, it corresponds to CH₃ bending or N-O asymmetric stretching in nitro groups. The peak at 1101 cm⁻¹ is commonly associated with C-wagging, and at 1037 cm⁻¹ is usually attributed to C-H in-plane bending in aromatic compounds⁴⁸.

Figure 2e exhibits the ¹H-NMR spectrum. ¹H-NMR titration was conducted to investigate the hydrogen bonding possibility between PCL/Gt and cefiderocol in synthesized nanofibers of CFD6, CFD12, and CFD24. Initially, the proton signals, in the range of 7–10 ppm, appeared broad and unclear. To address these challenges, ¹H-NMR spectra were recorded using a methanol-d₄: acetic acid-d₄ (1:1) solvent mixture. The dominant signals observed (3.31 ppm, 4.79 ppm, and 9.26 ppm) corresponded to the solvent peaks. Thus, the observed broad peak from 7 to 10 ppm mainly is the peak of solvents. The signals within the 1–5 ppm range are related to protons from both Gt and PCL^49,50. A specific signal at 2.9 ppm, corresponding to the lysine amino acid group in Gt, which is critical for hydrogen bonding, was identified. If increasing the weight percentage of cefiderocol had resulted in more hydrogen bonds, a corresponding chemical shift would be expected at 2.9 ppm. Here, no significant chemical shifts were observed for this signal across the different drug concentrations. The absence of shifts in the ¹H-NMR spectra for the three cefiderocol concentrations shows a lack of interaction between the drug and the nanofiber scaffolds. This is important in discussing the drug release properties of these nanofibers, which will be investigated later in this article.

Figure 2f presents the contact angle measurements of the ZX31PEO sample in comparison with those with nanofiber coating. The nanofiber layer in CFD6 enhances the surface hydrophilicity relative to that of ZX31PEO. However, as the cefiderocol concentration increases in CFD12 and CFD24, the surface becomes more hydrophobic than that in CFD6, indicating a correlation between the cefiderocol content and hydrophobicity.

Disk diffusion test

At the completion of the PEO and electrospinning coating processes, it is essential to evaluate the antibacterial properties incorporated within the nanofibers. Figure 3 presents the antibacterial disk diffusion test results of the bare alloy, ZX31PEO, CFD6, CFD12, and CFD24 samples (left) compared with those of the control sample (right) against E. coli ATCC 25922. The transparent-clear areas are the zones in which the bacteria could not grow. The bare ZX31 sample exhibited greater antibacterial properties than did the ZX31PEO sample, as indicated by the larger transparent inhibition zone around it (marked by a yellow dashed line). An exact disk diffusion diameter cannot be reported for the bare and ZX31PEO samples because bacterial inhibition in these cases does not result from the homogeneous release of an antibacterial agent but rather from patterns caused by chemical reactions. The increased corrosion resistance of the PEO-coated sample, compared with that of the bare alloy, affects its antibacterial effect. ZX31 underwent rapid corrosion, which contributed to bacterial killing, whereas the PEO coating on ZX31PEO provided increased corrosion resistance, thus reducing the bacterial inhibition by corrosion effects. The samples coated with antibiotic-laden nanofibers exhibited varying antibacterial effects depending on the cefiderocol content. With increasing CFD content, the disk diffusion diameters increase, with 26, 28, and 30 mm for CFD6, CFD12, and CFD24, respectively. Surprisingly, these values are still smaller than the diameter of 35 mm measured for the control sample. Even when an antibacterial agent is incorporated into a material, its controlled release is crucial, especially for potential applications in orthopedic surgeries. One of the key achievements of this research is the successful release of antibiotics.

Step 3: Immobilized phage on nanofibers; characterization and application

The disk diffusion test confirmed that the synthesized nanofibers were capable of releasing the antibiotic. The subsequent step involved immobilizing phages onto the nanofibers, and the success of this immobilization was examined using SEM and TEM analyses. Figure 4a exhibits the SEM image of CFD12 + Ø, as the representative of nanofibers incubated with phages. The sample prepared for SEM imaging was not washed so that the immobilized phages (marked in the blue square) and non-immobilized phages (depicted in a yellow circle) could be captured. The size of the non-immobilized phage is approximately 200 nm. The image indicates that this phage belongs to the Myoviridae family of phages^51,52. The blue square marks immobilized phages as small dots, similar to all the dots along the nanofibers in this figure. A comparison of this figure with Fig. 2a–c, at the same magnification, clearly indicates that the small dots are a large number of phages placed on the surface of the nanofibers. The difference between the size of the phage in the yellow circle and the size of the dots as the capsid of the phages is the distance between the non-immobilized phage and the immobilized phage.

Figure 4b shows a cross-sectional slice through a CFD12 + Ø sample, observed via TEM. The white area represents the cross-section of the nanofiber. The dark, big gray area and small black areas in the background of it, pointed by arrows, are the pores in the front and at the depth of the picture caused by the random placement of nanofibers. The small light circles in the dark gray areas are the phages attached to the fibers, as they were observed as small dots in the SEM images. The yellow arrows in Fig. 4b point to a phage at the capsid and tail parts. All the tails could not be captured by TEM because of their thin diameter. However, this image provides information regarding the intact immobilization of the phages without separation of capsids from the tails on the pores between the nanofibers. The release ability and antibacterial performance of the phages are important questions to explore next in this study.

Drug-releasing test

Once proper immobilization of the phages was confirmed, the method was considered complete, such that two antibacterial agents were integrated onto the initial PEO coating, which also modified the corrosion properties of the implant. Evaluating the release rate of these agents is crucial for determining the time scale over which this system can reliably protect the sample against bacterial infection. Figure 5 presents the results of the drug release tests in PBS for two groups of samples: CFD6, CFD12, and CFD24, containing only antibiotics, and CFD6 + Ø, CFD12 + Ø, and CFD24 + Ø, containing antibiotics and phages. For both groups with and without phages, the release curves cover each other, indicating that the release occurred regardless of the increase in the cefiderocol concentration. The release of the CFD6 was slightly sooner in the first few hours, as the CFD6 curve shows a shift to the left. However, its curve covers CFD12 and CFD24 from the 8^th hour. Hydrogen bonding is the most important factor contributing to crosslinking interactions between molecules in an electrospinning solution. However, the ¹H-NMR results shown in Fig. 2e proved that increasing the cefiderocol weight percentage does not increase the number of hydrogen bonds. El-Lababidi et al.³⁰ considered the pyrrolidinium group in the cefiderocol structure as the prevention recognition reason. Another significant observation in Fig. 5 is the difference between the release in the group with phages and the group without phages, such that the group with incubated phages shows a higher release rate. This means that phages accelerate the release.

Improved antibacterial effect in phage-antibiotic samples

The drug release test confirmed that both phages and cefiderocol were released within the defined time scale. The final and most important step, however, was to evaluate the antibacterial efficacy by comparing bacterial killing among samples containing both phages and antibiotics versus those containing either phages or antibiotics alone. Previous studies have investigated synergistic interactions between specific bacteriophages and antibiotics^2,3,43,53,54. Building on this knowledge, we selected combinations that had already demonstrated such effects. In particular, Fanaei Pirlar et al. ⁴³ reported a synergistic effect between the phage CUB-EC and the antibiotic cefiderocol. Thus, we incorporated this phage–antibiotic pair into our coating approach.

Figure 6 exhibits the antibacterial performance of the CFD6 + Ø, CFD12 + Ø, and CFD24 + Ø samples in contact with E. coli bacteria at 2, 4, 6, 8, 10, and 24 h. The bar diagrams show the survival of bacteria as CFUs. The empty space indicates no bacterial growth. The phage group, represented by Ø, shows the growth of bacteria treated only with phages over time. The ZX31PEO group exhibited the growth of bacteria during the mentioned time slots in contact with the ZX31PEO sample. The growth control group, shown by GC, which was not treated with antibiotics, phages, or ZX31PEO, presented a steady increase in CFUs over time, reflecting bacterial growth under normal conditions. During the first 2 h, the CFD6 + Ø, CFD12 + Ø, and CFD24 + Ø samples fully killed the bacteria. Although the CFD6, CFD12, and CFD24 samples exhibited bacterial growth during this period, the CFD6, CFD12, and CFD24 samples presented high, medium, and low bacterial growth, respectively, compared with each other. During the short period of 2 h, two factors could explain this decrease in the bacterial growth: the start of phage release and the thickness of the electrospun nanofibers. As depicted in Fig. 2, the thickness of the nanofibers increased with increasing amounts of cefiderocol. Therefore, CFD24 presented the highest effective surface along the nanofibers for contacting the bacteria. A greater effective surface, including greater amounts of cefiderocol, killed greater amounts of bacteria. This is also the case for CFD12 and CFD6.

After the 4^th hour, the CFD6 + Ø, CFD12 + Ø, and CFD24 + Ø samples showed no growth, which means that in these groups the bacteria are killed during the first 2 h. The CFD6, CFD12, and CFD24 samples exhibited almost equal growth of bacteria, meaning that the disposal of the nanofibers has started during this period, and drug release has started more effectively than the effective surface contact of bacteria with the nanofibers. However, the amount of the released drug is still not enough to completely kill bacteria, but compared to the growth control bar, the killing of bacteria has progressed.

The diagram of the behavior at the 6^th hour is similar to that at the 4^th hour. At the 8^th hour, the CFD12 and CFD24 samples killed the bacteria. However, the amount of released drug in the CFD6 sample was still not enough for full killing of the bacteria. This goal was also achieved for this sample and Ø groups, after the 10^th hour. The ZX31PEO group also completely killed the bacteria after the 10^th hour. The ratio of bacteria killed by this sample was not very different in the previous time slots, but between the 8th and 10th hours, a rapid killing of bacteria occurred. This means that the PEO layer protects the substrate during the first 8 h, but after this period, the coating layer cannot resist corrosion, and the substrate is exposed to the solution, and corrosion occurs quickly and at a high rate. Therefore, the bacteria that continued growing for 8 h were killed completely between the 8^th and 10^th hours. The group, including phages alone, Ø, was not strong enough to kill bacteria completely before it, which confirms the synergistic behavior of phages and antibiotics together. Between the 10^th and 24^th hour, no bacteria grew in the explored samples.

Considering using this strategy in the body of a patient, these results demonstrate that the CFD6 + Ø, CFD12 + Ø, and CFD24 + Ø samples began exhibiting antibacterial effects in the initial hours. Any surviving bacteria would be continuously exposed to cefiderocol released from the CFD6, CFD12, and CFD24 nanofibers. Additionally, the corrosion of the magnesium substrate further contributed to the antibacterial activity.

Investigating the reasons for synergy between the phage and cefiderocol

To elucidate why phage and cefiderocol eradicate bacteria more rapidly in combination than individually, it is essential to investigate their polarity both separately and together. Figure 7 is demonstrating the explanation of the synergic behavior of the phages and different concentrations of cefiderocol. Cefiderocol, with its various functional groups such as carbonyl groups, quaternary ammonium nitrogen, and hydrogen atoms bonded to electronegative atoms like nitrogen or oxygen, is a molecule exhibiting polarity^55,56. Phages also have an electric charge^57,58. The surface charge of phages is influenced by the surrounding environment, including the composition of nearby molecules and the characteristics of the medium^59,60. Figure 7 shows the schematic of bipolar momentum and respective electrical orbital cloud of phages and cefiderocol molecules before and after immobilization of phages. The capsids were immobilized on the nanofibers by Van der Waals forces between positive and negative charges of capsids and the polar sites of the cefiderocol. Due to the high positive charge in cefiderocol, after the immobilization, the negative charge of the electron clouds of capsids changes to higher negative charge, making the tails more positive. Thus, the tail with more positive electric charge could kill bacteria (E-coli as an example here) better. This enhances the antibacterial properties of the phages and explains the reason that phages combined with antibiotics exhibit stronger antibacterial effects than either alone.

To explain the shift to the left of the curves, including cefiderocol and phages in Fig. 5 compared to the samples without phages, Fig. 7 could clarify a suggested reason as follows. When the immobilized capsids have higher negative charges, their electron clouds apply a repulsive force to the negative electron cloud of CFD molecules, which have already become loose in contact with water and the ions of the medium. Fibers have solid phase, meaning that there is not big space to push the next cefiderocol molecules orbital cloud back. Therefore, its molecules get separated from the other molecules behind them, as shown by the blue dashed line in Fig. 7. This explains the reason that the drug release from the fibers with immobilized phages on them was accelerated in Fig. 5, compared to fibers without phages.

To reduce the risk of infection, antibiotic–phage formulations could be incorporated into pre-treated implants. The choice of antimicrobial agents should be tailored to the specific clinical context, and their release kinetics must be carefully evaluated and engineered prior to use. However, this study does not address the molecular polarization requirements of different antibiotics or the specificity of phage types. In conclusion, this work introduces an innovative antibacterial strategy for the prevention of implant infections, and tested components (cefiderocol, E. coli, and corresponding phage) can be easily replaced to address the desired pathogens.

Finally, the final coated product must protect against both Gram-negative and Gram-positive implant pathogens for clinical usage. Here, we only set out to find a coating method that works with antibiotics and bacteriophages. A final clinical product would use broad-spectrum antibiotics to be suitable for preventing both Gram-negative and Gram-positive implant pathogens, plus phage cocktails, as a mixture of different phages for different bacteria, targeting common implant pathogens, so the exact bacterium, phage, or antibiotic used in this study can be adapted to clinical needs.

Description of the substrate

An extruded ZX31 alloy was provided by Iran University of Science and Technology (IUST). It was cast in a cylindrical steel mold (200 mm length, 40 mm diameter), under argon shielding gas. To check the chemical composition, three samples were taken from the bottom, middle, and top sections of the cast cylinder, powdered, and analyzed using inductively coupled plasma (ICP) spectroscopy (Varian VISTA-PRO, Inc., Palo Alto, California, USA). The ICP results showed an average chemical composition of 95.85 ± 0.6 wt.-% magnesium, 3.12 ± 0.34 wt.-% zinc, and 1.07 ± 0.21 wt.% calcium. The extruded ZX31 alloy explored in this study is identical to that detailed in ref. ⁴⁵. The alloy was cut into 5 mm diameter, 1 mm thick disks via a wire-cutting method (Charmilles Robofil 240, Geneva, Switzerland). The samples prepared for the potentiodynamic polarization tests were cut into 10 × 10 × 5 mm³.

PEO coating process

To conduct the PEO process, a solution was prepared containing 10 g of Na₃PO₄·12H₂O (M = 380.13 g/mol, Merck, Alfa Aesar, Karlsruhe, Germany), 9 g of Na₂SiO₃·5H₂O (M = 212.14 g/mol, Sigma‒Aldrich, Alfa Aesar, Kandel, Germany), and 1 g of KOH (M = 56.11 g/mol, Merck, Saarbrücken, Germany) per liter. The electrical parameters were set as rectangular cathodic pulse mode for 300 s, and a voltage of 200 V by a power supply (Munk Nederland B.V., D400 G500/40 WRG-TFW, Kom., Nr.:246241, Hamm, Germany), and the software Visual Plating Controller (Munk Nederland B.V., Oirschot, Germany). A cooling system surrounded the stainless-steel bowl as the cathode, circulating cold water to keep the temperature of the electrolyte constant at ~ 25 °C. The abbreviation used for referring to the PEO-coated ZX31 samples in this study is ZX31PEO.

Potentiodynamic polarization measurements

To assess the biocorrosion properties of the ZX31 and ZX31PEO specimens, potentiodynamic polarization tests were performed (Ivium, Eindhoven, Netherlands). The square surface of 10 mm² was exposed to a phosphate-buffered saline (PBS) solution made from 10 OmniPur PBS tablets (Sigma-Aldrich, Darmstadt, Germany) dissolved in 1 L of distilled water, at 25 °C (solution composition is 140 mM NaCl, 10 mM phosphate buffer, and 3 mM KCl, pH 7.4 at 25 °C). To check the repeatability, three samples from each, ZX31 and ZX31PEO groups, were tested. The scan rate was set to 2.5 mV/s, covering a range from −200 to 0 mV. Tafel fitting analysis was used to determine the E_corr and i_corr values.

Electrospinning process

The electrospinning coating solution comprised Gt type A (Merck, Darmstadt, Germany), PCL (Mw = 80,000 g/mol, Sigma-Aldrich, Darmstadt, Germany), acetic acid (purity ≥ 99%, Merck, Darmstadt, Germany), formic acid (purity≥ 98%, Sigma–Aldrich, Darmstadt, Germany), and cefiderocol (Shionogi, Osaka, Japan). As previously reported in ref. ⁴⁶, the optimal approved concentrations of PCL and Gt, were 10 wt%, and 4 wt%, respectively. The employed concentrations of cefiderocol in this research are 6, 12, and 24 wt%, respectively. To prepare this solution, formic acid was mixed with acetic acid at a 2:1 ratio. PCL was stirred in the acid mixture for 1 h. Subsequently, Gt was added, and the solution was stirred for 30 min. Finally, different amounts of cefiderocol were added and stirred until the solution became transparent, indicating full dissolution. The electrospinning machine (FUG Bipolare high-voltage power supplies HCB, AIP Wild, Oberglatt, Switzerland) was set to a voltage of 20 kV, with a 10 cm distance between the collector and the syringe tip, and a flow rate (Syringe pump: TSE Systems 60 Series, Thuringia, East Germany) of 0.1 mL/h. The humidity was maintained below 40%. Since the collector in the electrospinning setup should be conductive, and PEO-coated samples are not so, the PEO-coated samples were affixed to an aluminum foil on the collector using double-sided tape. So the electrospinning machine recognized the aluminum foil as the collector, and the PEO-coated samples were covered by nanofibers supposed to be synthesized on the aluminum foil. Synthesizing duration was 2 hours for all samples. The abbreviations used for referring to the ZX31PEO samples covered by 6, 12, and 24 wt% cefiderocol-laden nanofibers in this study are CFD6, CFD12, and CFD24, respectively.

Characterization

A Quanta FEG 400 (FEI Company, Hillsboro, Oregon, USA) was used for SEM of the PEO coating and nanofibers before phage incubation. The specimens were gold sputtered for 20 s under an argon atmosphere at 10 Pa and 20 mA. The SEM parameters were set to high vacuum mode with secondary electron detection at 20.00 kV and a working distance of 12.2 mm. A cross-section of the ZX31PEO sample was made after embedding. The diameter of the nanofibers on the top surface was quantitatively analyzed over 100 random measurements of the thickness of the nanofibers via ImageJ (ImageJ software, National Institute of Health)⁶¹. For recognition of the functional groups in electrospun nanofibers, the layer of CFD 12 nanofibers, as the representative sample, was separated and explored by Fourier transform infrared (FTIR) spectroscopy. Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy (ATR-FTIR) was performed on a Bruker Vertex 70 device. (Bruker Optics GmbH & Co. KG, Ettlingen, Germany), covering 4000–400 cm⁻¹. To evaluate the possibility of hydrogen bonding, proton nuclear magnetic resonance (¹H-NMR) spectroscopy was performed on a Bruker Advance instrument (400 MHz) (Billerica, Massachusetts, USA) on CFD 6, CFD 12, and CFD 24 nanofiber layers, using deuterated solvents (Deutero GmbH, Kastellaun, Germany). Chemical shifts are referenced to residual impurities in the solvents.

The contact angle, made by a drop of water and explored surfaces of ZX31PEO, CFD 6, CFD 12, and CFD 24 samples, was measured via the sessile drop technique with a digital microscope (KEYENCE VHX-500, Keyence Deutschland GmbH, Neu-Isenburg, Germany) at room temperature. A 1 mL glass microsyringe filled with distilled water was used to dispense a 1.5 µL droplet onto the sample surface, with the piston controlled by a micrometer screw gauge. Each surface was tested five times, and the average contact angle was calculated. ImageJ software was used for analysis, and calculations were based on the model developed by Owens, Wendt, Rabel, and Kaelble⁶².

Disc diffusion test

The disc diffusion method was used to assess the antimicrobial activity of the bare alloy, ZX31PEO, and CFD 6, CFD 12, and CFD 24 samples based on the Clinical and Laboratory Standards Institute guideline (CLSI)⁶³. E. coli ATCC 25922 was chosen as the test organism, with a cefiderocol disk serving as the control. The set of tests for the examined groups of samples was repeated three times in the same conditions.

Immobilization of the phage

Phage CUB-EC, isolated by Fanaei Pirlar et al.⁴³, was utilized in this study as it has been shown to be specifically effective against E. coli ATCC 25922 (Labor Berlin – Charité Vivantes Services GmbH, Berlin, Germany). It was isolated by the enrichment method according to the process described in ref. ⁶⁴. For incubating phages onto the CFD6, CFD12, and CFD24 samples, they were exposed to a phage solution at a concentration of 10⁸ Plaque-Forming Units (PFU)/mL for 1 h, and addressed as CFD6 + Ø, CFD12 + Ø, and CFD24 + Ø, respectively in the following.

Observing phages on the nanofibers via SEM

To observe phages on the nanofibers via SEM, the samples were fixed in 2.5% glutaraldehyde (Serva, Heidelberg, Germany) in 0.1 M sodium cacodylate buffer (Serva, Heidelberg, Germany), post-fixed for 2 h with 2% osmium tetroxide (OsO₄), dehydrated through a graded ethanol series, and air-dried using hexamethyldisilazane (HMDS; Electron Microscopy Science, Hatfield, USA). The samples were carefully mounted on metal stubs. Prior to SEM analysis (GeminiSEM 300 with Image SP software, Carl Zeiss, Oberkochen, Germany), the samples were sputter-coated with gold-palladium (Sputter coater MED 020, Balzer, Bingen, Germany) and stored in a vacuum until examination. Images were captured at 7 kV using a secondary electron detector.

Observing phages on the nanofibers via transmission electron microscopy (TEM)

After incubation with the phage solution, for preparing TEM slices, the samples were fixed in 2.5% glutaraldehyde (Serva, Heidelberg, Germany) in 0.1 M sodium cacodylate buffer (Serva, Heidelberg, Germany), postfixed with 1% osmium tetroxide (Electron Microscopy Sciences, Hatfield, USA) and 0.8% potassium ferrocyanide II (Roth, Karlsruhe, Germany) in 0.1 M cacodylate buffer, dehydrated through a graded ethanol series, and embedded in Epon resin (Roth, Karlsruhe, Germany). Ultrathin sections (70 nm) were then cut from the hardened Epon blocks via an ultramicrotome (Leica, Wetzlar, Germany) with a diamond knife (Diatome, Nidau, Switzerland). These sections were collected on pioloform-coated copper grids (Plano, Wetzlar, Germany) and stained with lead citrate (Merck Millipore, Darmstadt, Germany). The slices were observed with a Zeiss Leo 906 TEM (Carl Zeiss, Oberkochen, Germany) equipped with a slow scan 2 K CCD camera (TRS, Moorenweis, Germany) at an accelerating voltage of 80 kV.

Drug release test

To examine the release of cefiderocol, a calibration curve was initially created via the Beer‒Lambert law with a PBS at pH 7.4 and 37 °C. Subsequently, a nanofibrous scaffolds of CFD6, CFD12, CFD24, and CFD6 + Ø, CFD12 + Ø, and CFD24 + Ø, samples were separated from the PEO-coated surfaces, and immersed in 20 mL of PBS at 25 °C. At predetermined intervals, every 2 h for the first group and every hour for the second group, 2 mL of the solution was removed from each sample for analysis, and 2 mL of fresh PBS was replaced to continue the release investigation. The amount of cefiderocol released into the PBS was monitored by measuring the UV absorbance at the peak wavelength of 260 nm via a UV spectrophotometer (Eppendorf BioPhotometer, Hamburg, Germany). The cumulative release percentage was then calculated. This process was repeated twice for all the samples to check the accuracy of the tests. Both repeats showed the same results.

Evaluating the antibacterial activity of released phage and antibiotic

To explore the antibacterial properties of CFD6 + Ø, CFD12 + Ø, and CFD24 + Ø, the samples were washed at least six times with a dip-wash method in sterile distilled water to rinse off non-immobilized phages. Bacterial suspension was prepared in cation-adjusted Muller-Hinton broth II (CAMHB) media. One mL of bacterial suspension at a concentration of 5 × 10⁵ CFU/mL was added to each well of a 24-well plate. The samples were subsequently placed into the wells. Bacterial counts were determined via colony-forming unit (CFU) counting at 2 h intervals. At each 2 h interval, 10 µL of the content from each well was taken and added to 990 µL of PBS. This mixture was centrifuged and washed three times to remove any remaining antimicrobial agents. After the final wash, 100 µL of the resuspended 1 mL solution was plated on tryptic soy agar (TSA) (Sigma‒Aldrich, Darmstadt, Germany) plates and incubated at 37 °C overnight for bacterial growth assessment. The tests were performed independently twice, each with duplicate samples (n = 4 in total).

1. Tarabichi, S. & Parvizi, J. Prevention of surgical site infection: a ten-step approach. Arthroplasty 5, 21 (2023).

   Article  PubMed  PubMed Central  Google Scholar 

2. Ma, T. et al. Incidence of periprosthetic joint infection after primary total knee arthroplasty shows significant variation: a synthesis of meta-analysis and bibliometric analysis. J. Orthop. Surg. 19, 649 (2024).

   Article  Google Scholar 

3. Ayoade, F., Li, D., Mabrouk, A. & Todd, J. R. Periprosthetic Joint Infection. in StatPearls [Internet] (StatPearls Publishing, 2023).

4. Savvidou, O. D. et al. Efficacy of antimicrobial coated orthopaedic implants on the prevention of periprosthetic infections: a systematic review and meta-analysis. J. Bone Jt. Infect. 5, 212–222 (2020).

   Article  PubMed  PubMed Central  Google Scholar 

5. Stewart, P. S. & William Costerton, J. Antibiotic resistance of bacteria in biofilms. The Lancet 358, 135–138 (2001).

   Article  CAS  Google Scholar 

6. Zalavras, C. G., Patzakis, M. J. & Holtom, P. Local Antibiotic Therapy in the Treatment of Open Fractures and Osteomyelitis. Clin. Orthop 427, 86–93 (2004).

   Article  Google Scholar 

7. Alt, V. Antimicrobial coated implants in trauma and orthopaedics–A clinical review and risk-benefit analysis. Injury 48, 599–607 (2017).

   Article  PubMed  Google Scholar 

8. Staiger, M. P., Pietak, A. M., Huadmai, J. & Dias, G. Magnesium and its alloys as orthopedic biomaterials: A review. Biomaterials 27, 1728–1734 (2006).

   Article  CAS  PubMed  Google Scholar 

9. Nasr Azadani, M., Zahedi, A., Bowoto, O. K. & Oladapo, B. I. A review of current challenges and prospects of magnesium and its alloy for bone implant applications. Prog. Biomater. 11, 1–26 (2022).

   Article  CAS  PubMed  PubMed Central  Google Scholar 

10. Lin, Z., Sun, X. & Yang, H. The Role of Antibacterial Metallic Elements in Simultaneously Improving the Corrosion Resistance and Antibacterial Activity of Magnesium Alloys. Mater. Des. 198, 109350 (2021).

    Article  CAS  Google Scholar 

11. Xie, K. et al. Additively manufactured biodegradable porous magnesium implants for elimination of implant-related infections: An in vitro and in vivo study. Bioact. Mater. 8, 140–152 (2022).

    CAS  PubMed  Google Scholar 

12. Hornberger, H., Virtanen, S. & Boccaccini, A. R. Biomedical coatings on magnesium alloys – A review. Acta Biomater 8, 2442–2455 (2012).

    Article  CAS  PubMed  Google Scholar 

13. Gu, X. et al. Corrosion of, and cellular responses to Mg–Zn–Ca bulk metallic glasses. Biomaterials 31, 1093–1103 (2010).

    Article  CAS  PubMed  Google Scholar 

14. Barati Darband, G. H., Aliofkhazraei, M., Hamghalam, P. & Valizade, N. Plasma electrolytic oxidation of magnesium and its alloys: Mechanism, properties and applications. J. Magnes. Alloys 5, 74–132 (2017).

    Article  CAS  Google Scholar 

15. Lashkarara, S., Fazlali, A., Ghaseminezhad, K., Fleck, C. & Salavati, M. Mechanism of plasma electrolytic oxidation in Mg3ZnCa implants: a study of double-layer formation and properties through nanoindentation. Sci. Rep. 14, 7380 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

16. Sinkevitch, Y. V. Conceptual model of commutation mechanism for electric conductivity of vapor-gas envelope in electro-impulse polishing mode. Sci. Tech. 15, 407–414 (2016).

    Google Scholar 

17. Fattah-alhosseini, A., Babaei, K. & Molaei, M. Plasma electrolytic oxidation (PEO) treatment of zinc and its alloys: A review. Surf. Interfaces 18, 100441 (2020).

    Article  CAS  Google Scholar 

18. Kamil, M. P., Kaseem, M. & Ko, Y. G. Soft plasma electrolysis with complex ions for optimizing electrochemical performance. Sci. Rep. 7, 44458 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

19. Lu, X. et al. Influence of SiO2 Particles on the Corrosion and Wear Resistance of Plasma Electrolytic Oxidation-Coated AM50 Mg Alloy. Coatings 8, 306 (2018).

    Article  Google Scholar 

20. Ma, C. et al. Anticorrosive non-crystalline coating prepared by plasma electrolytic oxidation for ship low carbon steel pipes. Sci. Rep. 10, 15675 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

21. Polo, T. O. B. et al. Plasma Electrolytic Oxidation as a Feasible Surface Treatment for Biomedical Applications: an in vivo study. Sci. Rep. 10, 10000 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

22. Chen, X., Zhou, J., Qian, Y. & Zhao, L. Antibacterial coatings on orthopedic implants. Mater. Today Bio 19, 100586 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

23. Bhardwaj, N. & Kundu, S. C. Electrospinning: A fascinating fiber fabrication technique. Biotechnol. Adv. 28, 325–347 (2010).

    Article  CAS  PubMed  Google Scholar 

24. Xue, J. et al. Fabrication and evaluation of electrospun PCL–gelatin micro-/nanofiber membranes for anti-infective GTR implants. J Mater Chem B 2, 6867–6877 (2014).

    Article  CAS  PubMed  Google Scholar 

25. Nazari, S., Naeimi, M., Rafienia, M. & Monajjemi, M. Fabrication and Characterization of 3D Nanostructured Polycaprolactone-Gelatin/Nanohydroxyapatite-Nanoclay Scaffolds for Bone Tissue Regeneration. J. Polym. Environ. https://doi.org/10.1007/s10924-023-02966-z(2023).

26. Ajmal, G. et al. Biomimetic PCL-gelatin based nanofibers loaded with ciprofloxacin hydrochloride and quercetin: A potential antibacterial and anti-oxidant dressing material for accelerated healing of a full thickness wound. Int. J. Pharm. 567, 1–12 (2019).

    Article  Google Scholar 

27. Abdul Khodir, W. K. W., Abdul Razak, A. H., Ng, M. H., Guarino, V. & Susanti, D. Encapsulation and Characterization of Gentamicin Sulfate in the Collagen Added Electrospun Nanofibers for Skin Regeneration. J. Funct. Biomater. 9, 36 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

28. Aydemir Sezer, U., Arslantunali, D., Aksoy, E. A., Hasirci, V. & Hasirci, N. Poly(ε-caprolactone) composite scaffolds loaded with gentamicin-containing β-tricalcium phosphate/gelatin microspheres for bone tissue engineering applications. J. Appl. Polym. Sci 131, app.40110 (2014).

    Article  Google Scholar 

29. Bakhsheshi-Rad, H. R. et al. Drug delivery and cytocompatibility of ciprofloxacin loaded gelatin nanofibers-coated Mg alloy. Mater. Lett. 207, 179–182 (2017).

    Article  CAS  Google Scholar 

30. Aoki, T. et al. Cefiderocol (S-649266), A new siderophore cephalosporin exhibiting potent activities against Pseudomonas aeruginosa and other gram-negative pathogens including multi-drug resistant bacteria: Structure activity relationship. Eur. J. Med. Chem. 155, 847–868 (2018).

    Article  CAS  PubMed  Google Scholar 

31. Butler, M. S., Blaskovich, M. A. & Cooper, M. A. Antibiotics in the clinical pipeline at the end of 2015. J. Antibiot. ((Tokyo)) 70, 3–24 (2017).

    Article  CAS  PubMed  Google Scholar 

32. Crémet, L. et al. Orthopaedic-implant infections by Escherichia coli: Molecular and phenotypic analysis of the causative strains. J. Infect. 64, 169–175 (2012).

    Article  PubMed  Google Scholar 

33. Crémet, L. et al. Innate immune evasion of Escherichia coli clinical strains from orthopedic implant infections. Eur. J. Clin. Microbiol. Infect. Dis. 35, 993–999 (2016).

    Article  PubMed  Google Scholar 

34. Polascik, B. A. et al. 976. E coli Periprosthetic Joint Infections: Poor Infection Clearance at One Year. Open Forum Infect. Dis. 9, ofac492.818 (2022).

    Article  PubMed Central  Google Scholar 

35. Pirnay, J.-P. et al. Personalized bacteriophage therapy outcomes for 100 consecutive cases: a multicentre, multinational, retrospective observational study. Nat. Microbiol. 9, 1434–1453 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

36. Manohar, P., Loh, B., Nachimuthu, R. & Leptihn, S. Phage-antibiotic combinations to control Pseudomonas aeruginosa–Candida two-species biofilms. Sci. Rep. 14, 9354 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

37. Wintachai, P. et al. Enhanced antibacterial effect of a novel Friunavirus phage vWU2001 in combination with colistin against carbapenem-resistant Acinetobacter baumannii. Sci. Rep. 12, 2633 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

38. Mukhopadhyay, S. et al. Sequential treatment effects on phage–antibiotic synergistic application against multi-drug-resistant Acinetobacter baumannii. Int. J. Antimicrob. Agents 62, 106951 (2023).

    Article  CAS  PubMed  Google Scholar 

39. Alič, Š et al. Newly Isolated Bacteriophages from the Podoviridae, Siphoviridae, and Myoviridae Families Have Variable Effects on Putative Novel Dickeya spp. Front. Microbiol. 8, 1870 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

40. Fokine, A. & Rossmann, M. G. Molecular architecture of tailed double-stranded DNA phages. Bacteriophage 4, e28281 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

41. O’Connell, L., Marcoux, P. R. & Roupioz, Y. Strategies for Surface Immobilization of Whole Bacteriophages: A Review. ACS Biomater. Sci. Eng. 7, 1987–2014 (2021).

    Article  PubMed  Google Scholar 

42. Eskenazi, A. et al. Combination of pre-adapted bacteriophage therapy and antibiotics for treatment of fracture-related infection due to pandrug-resistant Klebsiella pneumoniae. Nat. Commun. 13, 302 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

43. Pirlar, R. F., Halili, N., Travnik, T., Trampuz, A. & Karbysheva, S. In vitro activity of cefiderocol against Gram-negative aerobic bacilli in planktonic and biofilm form–alone and in combination with bacteriophages. Sci. Rep. 15, 17105 (2025).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

44. Vaghefinazari, B. et al. Exploring the corrosion inhibition mechanism of 8-hydroxyquinoline for a PEO-coated magnesium alloy. Corros. Sci. 203, 110344 (2022).

    Article  CAS  Google Scholar 

45. Lu, Y. Microstructure and degradation behaviour of Mg-Zn(-Ca) alloys. (University of Birmingham, 2014).

46. Ghaseminezhad, K., Zare, M., Lashkarara, S., Yousefzadeh, M. & Mohandesi, J. A. Fabrication of althea officinalis loaded electrospun nanofibrous scaffold for potential application of skin tissue engineering. J. Appl. Polym. Sci. 137, 1–9 (2020).

    Article  Google Scholar 

47. Muyonga, J. H., Cole, C. G. B. & Duodu, K. G. Fourier transform infrared (FTIR) spectroscopic study of acid soluble collagen and gelatin from skins and bones of young and adult Nile perch (Lates niloticus). Food Chem. 86, 325–332 (2004).

    Article  CAS  Google Scholar 

48. Kumar, R., Kamakshi, Kumar, M. & Awasthi, K. Functionalized Pd-decorated and aligned MWCNTs in polycarbonate as a selective membrane for hydrogen separation. Int. J. Hydrog. Energy 41, 23057–23066 (2016).

    Article  CAS  Google Scholar 

49. Kerman, I., Toppare, L., Yilmaz, F. & Yagci, Y. Thiophene Ended ϵ-Caprolactone Conducting Copolymers and their Electrochromic Properties. J. Macromol. Sci. Part A 42, 509–520 (2005).

    Article  Google Scholar 

50. Li, X. et al. 3D Culture of Chondrocytes in Gelatin Hydrogels with Different Stiffness. Polymers 8, 269 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

51. Berkson, J. D. et al. Phage-specific immunity impairs efficacy of bacteriophage targeting Vancomycin Resistant Enterococcus in a murine model. Nat. Commun. 15, 2993 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

52. Li, F. et al. High-resolution cryo-EM structure of the Pseudomonas bacteriophage E217. Nat. Commun. 14, 4052 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

53. Comeau, A. M., Tétart, F., Trojet, S. N., Prère, M.-F. & Krisch, H. M. Phage-Antibiotic Synergy (PAS): β-Lactam and Quinolone Antibiotics Stimulate Virulent Phage Growth. PLOS ONE 2, e799 (2007).

    Article  PubMed  PubMed Central  Google Scholar 

54. Phage-antibiotic combination is a superior treatment against Acinetobacter baumannii in a preclinical study – PubMed. https://pubmed.ncbi.nlm.nih.gov/35537278/.

55. Soriano, A. & Mensa, J. Mechanism of action of cefiderocol. Rev. Esp. Quimioter. 35, 16–19 (2022).

    Article  PubMed  PubMed Central  Google Scholar 

56. Domingues, S., Lima, T., Saavedra, M. J. & Da Silva, G. J. An Overview of Cefiderocol’s Therapeutic Potential and Underlying Resistance Mechanisms. Life 13, 1427 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

57. Zhu, J. et al. Design of bacteriophage T4-based artificial viral vectors for human genome remodeling. Nat. Commun. 14, 2928 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

58. Zhou, Y., Marar, A., Kner, P. & Ramasamy, R. P. Charge-Directed Immobilization of Bacteriophage on Nanostructured Electrode for Whole-Cell Electrochemical Biosensors. Anal. Chem. 89, 5734–5741 (2017).

    Article  CAS  PubMed  Google Scholar 

59. Heffron, J. & Mayer, B. K. Virus Isoelectric Point Estimation: Theories and Methods. Appl. Environ. Microbiol. 87, e02319–20 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

60. Szermer-Olearnik, B. et al. Aggregation/dispersion transitions of T4 phage triggered by environmental ion availability. J. Nanobiotechnol. 15, 32 (2017).

    Article  Google Scholar 

61. Abràmoff, M. D., Magalhaes, P. J. & Ram, S. J. Image Processing with ImageJ. Biophotonics International 11, 36–42 (2004).

    Google Scholar 

62. Owens, D. K. & Wendt, R. C. Estimation of the surface free energy of polymers. J. Appl. Polym. Sci. 13, 1741–1747 (1969).

    Article  CAS  Google Scholar 

63. Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing: Edition 31: Clinical and Laboratory Standards Institute Wayne, PA; 2021.

64. Manohar, P., Tamhankar, A. J., Lundborg, C. S. & Ramesh, N. Isolation, characterization and in vivo efficacy of Escherichia phage myPSH1131. PLOS ONE 13, e0206278 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Author notes

1. These authors contributed equally: S. Lashkarara, R. Fanaei Pirlar.

Authors and Affiliations

1. Technische Universität Berlin, Faculty III Process Sciences, Institute of Materials Science and Technology, Fachgebiet Werkstofftechnik/Chair of Materials Science & Engineering, Berlin, Germany

   S. Lashkarara, M. Salavati & C. Fleck

2. Charité—Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Center for Musculoskeletal Surgery, Berlin, Germany

   R. Fanaei Pirlar & A. Trampuz

3. Friedrich Alexander University Erlangen Nürnberg, Department of Chemistry and Pharmacy, Erlangen, Germany

   M. Akhgar

4. School of Medicine, Faculty of Health, Queensland University of Technology, Brisbane, QLD, Australia

   A. Trampuz

5. Royal Brisbane and Women’s Hospital, Herston, QLD, Australia

   A. Trampuz

Contributions

S.L. and R.F.P. conceptualized the study. S.L. and R.F.P. developed the methodology and conducted the validation alongside C.F., A.T., M.S., and M.A.; Formal analysis and investigation were carried out by S.L. and R.F.P.; Resources were provided by S.L., A.T., M.A., and C.F.; S.L., R.F.P., and M.A. curated the data. The original draft was prepared by S.L., R.F.P., and M.A.; Review and editing by C.F., S.L., R.F.P., and M.A.; Supervision was overseen by C.F.; Project administration by S.L., R.F.P., and C.F.; R.F.P. contributed to this study as co-first author. Funding acquisition was managed by C.F., S.L., and A.T. All authors have read and approved the manuscript.

Corresponding author

Correspondence to S. Lashkarara.

Competing interests

The authors declare no competing interests.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions