Stabilization of Niaouli essential oil-loaded gelatin nanofibers: characterization and in vitro evaluation

Abstract

Niaouli essential oil (NEO), obtained from Melaleuca quinquenervia leaves, is a plant-derived active compound with high biological activity properties; however, its direct use is limited due to its hydrophobicity, volatility, and instability. In this study, biologically active and mechanically enhanced nanofiber-based drug delivery systems were developed by directly incorporating NEO into gelatin (G)-based electrospun nanofibers and enhancing their stability through chemical crosslinking with glutaraldehyde (GTA) vapor. In this context, electrospun nanofibers loaded with NEO concentrations of 1% (GN1), 5% (GN5), and 10% v/v (GN10). They were exposed to the vapor generated by the GTA solution for different durations (10, 30 min, 1 h, and 3 h), temperature settings (25 and 50 °C), and two different volumes (50 and 100 µL). Morphological analysis and water solubility tests were performed to determine the optimal crosslinking conditions. Subsequently, both GN1 and GN5 NEO-loaded nanofibers, crosslinked for 1 h and 3 h were comprehensively characterized in terms of morphology, chemical structure, wettability, mechanical strength, thermal stability, in vitro drug release behavior, and antimicrobial activity. Following 3 h of crosslinking, GN1_C3 nanofibers exhibited a rapid initial release (92% in ~ 120 h), while GN5_C3 nanofibers provided a slower, sustained release (75% in ~ 240 h). Biological evaluations using MTT assay, fluorescence microscopy, and SEM confirmed high cytocompatibility and effective support of L929 fibroblast adhesion and proliferation. Strong antibacterial activity was observed against S. aureus and E. coli for GN5_C3. To the best of our knowledge, this is the first study reporting the direct incorporation of NEO into electrospun nanofibers. The fabricated NEO-loaded and crosslinked nanofibers represent a platform for tissue engineering and antimicrobial applications due to their high biological and antimicrobial properties.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-54222-8.

Introduction

Essential oils (EOs) with high biological activity are complex mixtures synthesized by medicinal and aromatic plants for defensive or signaling purposes¹. While numerous EOs (approximately 3,000) have been identified to date, only 300 of them are reportedly used commercially. This demonstrates that a large number of EOs with potential applications remain to be discovered and evaluated^2,3. Among these, Niaouli essential oil (NEO) is one of the oils whose full potential remains largely unexplored. Obtained from the Melaleuca quinquenervia plant, native to Northern Australia, NEO is rich in monoterpenes and sesquiterpenes, and its most important functional component, 1,8-cineole (≈ 55%), has been reported and confirmed to be responsible for NEO’s therapeutic activity^4,5.

NEO has been used in traditional medicine since ancient times for various purposes (e.g., aromatherapy, antiseptic applications, and wound healing). Recently, NEO has become highly popular for its antibacterial^6–8, antifungal³, anti-inflammatory⁹, and antioxidant effects^8,10. The inherent properties of NEO, such as high volatility, instability, rapid degradation, and hydrophobicity, limit its direct use in practical applications. To overcome these limitations, encapsulation of NEO into a controlled drug delivery system is the most feasible solution to both minimize potential side effects and achieve more effective therapeutic outcomes^11,12. Specifically, encapsulation into electrospun nanofibers, spray drying, coacervation, co-crystallization or ionic gelation are among the widely applied strategies in the food, polymer, and pharmaceutical industries^13,14.

Nanofibers are advanced materials with very wide application areas and ultrathin diameters typically ranging from 50 to 500 nm¹⁵. Their high surface-to-volume ratio, tunable porosity, and ability to mimic the structural resemblance of the extracellular matrix (ECM) not only enhance cell adhesion and proliferation but also improve encapsulation efficiency and controlled release of bioactive compounds. These properties make nanofibers an ideal platform for stabilizing EOs and enabling their effective application in tissue engineering, wound healing, and other biomedical contexts¹⁶. Although various technologies are used for nanofiber production, electrospinning has emerged as the most widely used method due to its practicality, affordability, and versatility in producing fibers at the nanometer to micrometer scale. This technique enables the production of nanofibers from both natural and synthetic polymers, enabling the encapsulation of a wide variety of bioactive compounds^17,18. Among natural polymers, gelatin (G) is one of the most promising candidates due to its high biocompatibility, biodegradability, structural similarity to ECM, and amenability to chemical modifications^19,20. The properties described for G are critical in delivery systems, particularly nanofiber-based platforms, where controlled drug release, enhanced stability of encapsulated compounds, and improved cellular interactions are important^21–23.

Several studies have reported the incorporation of EOs into G nanofibers and their subsequent stabilization. For example, Zhou et al. (2020) incorporated angelica EO into G nanofibers for active food packaging applications, and loaded EO conferred antimicrobial and/or antioxidant properties to the nanofibers. In another study, orange EO was incorporated into G nanofibers, and the nanofibers were crosslinked with tannic acid. According to result of the study, orange EO was effectively encapsulated within the nanofiber matrix and performed controlled release. Additionally, the crosslinking process enhanced the nanofibers’ resistance to water solubility¹⁴. In another study, chitosan/G-based nanofibers loaded with oregano EO were developed using a nozzleless electrospinning method and tested in a food matrix application to reduce nitrite levels in sausages. The results exhibited that the nanofibers were stable and durable, and the antibacterial activity of the added oregano oil resulted in a reduction in nitrite levels in sausages²⁴. It is possible to increase the number of examples of G-based nanofibers loaded with EOs that have been produced for different purposes in the literature. However, the main challenge with G-based nanofibers is their rapid degradation in aqueous environments due to their high hydrophilicity²⁵. To overcome this limitation, crosslinking strategies have been developed to increase the mechanical strength of G nanofibers, their resistance to rapid degradation, and their structural integrity^14,19.

A wide range of physical and chemical approaches have been investigated for the crosslinking of collagen-based materials. While physical techniques such as dehydrothermal treatment and UV irradiation were initially employed, these methods generally showed lower efficiency compared to chemical strategies. Among chemical crosslinkers, compounds such as genipin, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC)²⁶, hexamethylene diisocyanate (HMDI)²⁷, dextran dialdehyde, and glutaraldehyde (GTA) have been extensively used to modify G for biomedical applications. In particular, GTA remains the most widely used crosslinker due to its high efficiency in stabilizing collagen and gelatin scaffolds^18,28. GTA-mediated crosslinking helps maintain the biocompatibility, mechanical strength, and flexibility of G-based materials. Additionally, GTA is cost-effective, readily available, and allows crosslinking in relatively short processing times, making it a practical option for stabilizing collagenous and gelatinous scaffolds in tissue engineering applications^29,30. Although GTA is the most suitable chemical crosslinking agent, it is toxic and can persistent within the nanofibers. Therefore, alternative approaches such as vapor-phase crosslinking or vacuum-drying have been developed for reducing toxicity and controlling its concentration¹⁸.

Various examples of G-based nanofibers crosslinked with GTA vapor exist in the literature. For example, Pirmoradian et al. (2023) produced nanofiber membranes by blending G with a copolymer and crosslinking them using GTA vapor. The crosslinking process increased the mechanical strength and stability of the membranes, and they remained intact for up to 30 days³¹. In another study, PVA/G based nanofibers were exposed to GTA vapor for different durations, and it was observed that the mechanical strength of the membranes increased up to a certain point, while excessive crosslinking led to brittleness. In addition, it was observed that the crosslinking process affected the degradation behavior and active compound release kinetics of the nanofibers. Crosslinked fibers exhibited more prolonged and controlled release compared to non-crosslinked fibers³². In a study by Chen et al.(2022), nanofiber non-woven fabrics were produced using a blend of feather keratin (FK) and G as a waste protein source via electrospinning, and the fibers were exposed to GTA vapor for different durations. The results showed that the nanofibers became more stable and compact, and they supported rapid proliferation of human foreskin fibroblast cells³³.

According to our comprehensive literature review, studies on the incorporation of NEO into G-based nanofibers and their crosslinking with GTA vapor are limited, and this topic has not been sufficiently addressed in the literature. Although some studies have examined the integration of NEO matrices^8,34–36, research on its direct loading into nanofibers via electrospinning is almost nonexistent. This study holds originality as it involves the direct incorporation of NEO into gelatin nanofibers and their stabilization through GTA vapor crosslinking.

The aim of this study was to develop a potential wound dressing or tissue engineering scaffold based on NEO-loaded gelatin nanofibers with improved mechanical properties through GTA vapor crosslinking. Furthermore, the study aimed to evaluate the antimicrobial activity, biocompatibility, and in vitro release behavior of the developed nanofibers to assess their suitability for biomedical applications. In the study, G nanofibers containing different concentrations of NEO were produced, and crosslinking was performed under various temperatures, durations, and GTA vapor conditions. The crosslinked nanofibers were examined in terms of morphological properties and wettability, and selected samples were characterized in detail. The physicochemical properties of the polymeric solutions used for nanofiber fabrication, including viscosity, electrical conductivity, and surface tension, were evaluated. Additionally, the nanofibers were characterized in terms of their chemical, morphological, mechanical, and thermal properties, along with their hydrophilic/hydrophobic behavior and in vitro drug release profile. Finally, the developed NEO-loaded nanofibers were subjected to antimicrobial testing against S. aureus and E. coli using the disk diffusion method, and their in vitro biocompatibility was comprehensively evaluated using the L929 fibroblast cell line.

Results

The production of the nanofibers was initially evaluated by directly viewing the electrospinning process, including jet stability and fibre deposition on the collector. The G-based solutions had appropriate viscosity and spinnability, enabling the formation of a stable Taylor cone under the applied voltage (Table SI1). Continuous and uniform jets were produced for formulations containing 1–5% (v/v) NEO, resulting in the efficient deposition of homogeneous nanofibrous mats. However, the addition of 10% (v/v) NEO adversely affected the electrospinning process, leading to jet instability and non-uniform fibre formation, likely due to excessive oil content and reduced chain entanglement (Table SI2). Consequently, subsequent experiments were conducted using NEO-loaded nanofibers at 1% and 5% (v/v) concentrations. After electrospinning process, the nanofibers were subjected to GTA vapor crosslinking under various conditions (Table SI3–SI10). It was observed that increasing both GTA amount and temperature led to noticeable shrinkage of the nanofibers and deterioration of the morphological structure. Therefore, the most suitable crosslinking condition was determined as 50 µL GTA at 25 °C for 1 h, and the crosslinking time was further extended to 3 h for comparison. The crosslinked nanofibers were subsequently used for further characterization analyses.

Physical properties characterization

The physicochemical properties (viscosity, surface tension and electrical conductivity) of both G and NEO-incorporated solutions were measured and results are summarized at Table 1. Surface tension is an important parameter in electrospinning processes, particularly in jet formation and final fiber morphology. High surface tension increases the surface energy of liquid droplets, which promotes bead formation. Conversely, reduced surface tension leads to the production of smoother and more uniformly distributed nanofibers. Previous studies on G-based report the surface tension of G polymer solutions typically ranges between 20 and 40 mN/m^37–39. In this study, the surface tension of the G solution was measured as 35.30 mN/m. Addition of 1% (v/v) NEO (GN1) did not significantly change the surface tension, while 5% (v/v) NEO (GN5) led to a slight decrease in surface tension.

Table 1.

Physicochemical properties of G and NEO-loaded solutions.

Polymer	Surface Tension (mN/m)	Electrical Conductivity (mS/cm)	Viscosity at 100 s− 1 (mPa.s)
G	35.30 ± 1.51	5.52 ± 2.27	99.28
GN1	35.09 ± 0.21	5.64 ± 1.50	87.35
GN5	32.79 ± 0.33	5.27 ± 1.65	92.46

Adequate conductivity is crucial for the electrospinning jet to be stable, continuous and well-controlled. Formation of a stable jet depends on increased charge transfer with high solution conductivity. However, low conductivity can cause problems in fiber production³⁹. In this study, the addition of 1% and 5% (v/v) NEO did not significantly affect the electrical conductivity, and only minor variations were observed among the samples (Table 1).

Viscosity is a fundamental factor in electrospinning and directly affects the solution flow, stability, and fiber morphologies. If the viscosity is too low, beads and irregularities can form in the fibers; if it is too high, jet stability is impaired. Therefore, maintaining an optimal viscosity range is essential to obtain uniform, bead-free, and well-formed nanofibers⁴⁰. The rheological properties of G solutions, consisting of 18% (w/v) G dissolved in a mixture of formic acid: acetic acid (70:30, v/v) and NEO (1% and 5% v/v) added G solutions were determined using steady-shear rheometry in a shear rate range of 0–500 1/s (Fig. 1a, b). The viscosity values of all solutions at a shear rate of 100 s⁻¹ are summarized in Table 1.

Fig. 1.

Rheologic behavior of G and GN solutions (a) Shear stress–shear rate (b) apparent viscosity examine with shear rate.

Morphological analyses of the nanofibers

As a result of the optimization studies, nanofibers loaded with 1% (v/v) and 5% (v/v) NEO exhibited homogeneous and uniformly distributed morphological features, whereas those loaded with 10% (v/v) NEO displayed non-uniform morphology (Table SI2). For 1% and 5% (v/v) NEO-loaded nanofibers, the optimal electrospinning parameters were identified as a flow rate of 1 mL/h, a high voltage of 24 kV, and a working distance of 10 cm. Under these conditions, the nanofibers were exposed to vapor-phase crosslinking using 50 µL of 50% aqueous GTA solution at room temperature for different durations (10 min, 30 min, 1 h, and 3 h), and crosslinking optimization was evaluated accordingly.

In this study, SEM images and fiber diameter distribution graphs of non-crosslinked and GTA vapor-crosslinked nanofibers (for 1 h and 3 h) at 25 °C are presented in Fig. 2. The average diameter of the non-crosslinked G nanofibers was measured as 273 ± 59 nm (Fig. 2a). By the addition of 1% (v/v) NEO, the fiber diameter slightly increased to 281 ± 84 nm (Fig. 2d), though this change was not statistically significant. When the NEO content was increased to 5% (v/v), the fiber diameter decreased to 253 ± 71 nm (Fig. 2g).

Fig. 2.

SEM images and fiber diameter distribution of electrospun nanofibers G (a) and after crosslinking for: G_C1 (b), G_C3 (c), and GN1 (d), after crosslinking for GN1_C1 (e), GN1_C3 (f), and GN5 (g), after crosslinking for GN5_C1 (h), GN5_C3 (i). Calculated by averaging the diameter of 100 fibers. Scale bar: 1 μm.

FTIR spectroscopy

The FTIR spectra of both non-crosslinked (G) and crosslinked G (G_C1, G_C3) nanofibers, as well as nanofibers loaded with NEO, are presented in Fig. 3, covering the spectral range of 400–4000 cm⁻¹. A broad absorption band observed around 3285–3296 cm⁻¹ in all G based samples corresponds to the overlapping O–H and N–H stretching vibrations, indicative of hydrogen bonding and the presence of amino groups in the gelatin structure. In the spectrum of the G, characteristic amide bands were observed: amide I at 1641 cm⁻¹ (C = O stretching), amide II in the 1532 cm⁻¹ (N–H bending), and amide III at approximately 1229.34 cm⁻¹ (C–N stretching). Additionally, bands between 1405 and 1450 cm⁻¹ are attributed to the –CH₂– bending vibrations of aliphatic chains, which are typical of the gelatin backbone. In the NEO, peaks were recorded at 1374.87, 1215.56, 1080, and 985 cm⁻¹. These bands are associated with the main active compound in NEO, 1,8-cineole. The bands at 1080 cm⁻¹ correspond to C–O stretching vibrations, while the peak at 1374 cm⁻¹ is attributed to symmetric bending of methyl (CH₃) groups. Peaks observed at 1215 cm⁻¹ are characteristic of asymmetric and symmetric C–O–C stretching, respectively⁴¹.

Fig. 3.

FTIR spectra of non-crosslinked G, GN1, GN5 nanofibers, and NEO and fibers after crosslinking for: G_C1, G_C3, GN1_C1, GN1_C3, GN5_C1, GN5_C3.

The FTIR spectra of G-based and NEO-loaded nanofibers (GN1 and GN5) showed similar characteristic bands, indicating that NEO incorporation did not significantly alter the chemical structure of the gelatin matrix. This suggests that NEO was physically encapsulated within the nanofibers rather than forming covalent bonds, consistent with previous studies^42,43. When comparing the crosslinking durations of the nanofibers, it is observed that the characteristic peaks exhibit a redshift as the crosslinking time increases. The decrease in the intensity of the Amide I peaks, which is indicated by red arrows in the figures, serves as evidence for this trend³³. These findings indicate that both pure gelatin (G) and NEO-loaded gelatin nanofibers were crosslinked by GTA vapor, leading to a reduction in the number of free amino and hydroxyl groups. This phenomenon can be attributed to the fact that excess GTA disrupts the triple-helix structure of gelatin, thereby increasing the density of hydrophilic functional groups and enhancing hydrogen-bonding interactions²⁹.

Thermal properties

DSC (Differential Scanning Calorimetry) analysis was used to investigate the thermal behavior of nanofiber patches. All nanofiber samples exhibited endothermic transitions in the temperature ranges of 80–90 °C and 250–270 °C, attributed to moisture evaporation and structural denaturation, respectively (Fig. 4). Non-crosslinked gelatin nanofibers exhibited two main endothermic transitions at approximately 73 °C and 260 °C, corresponding to the evaporation of bound water/partial disruption of the triple-helix structure and thermal degradation of the gelatin backbone, respectively^44,45. The addition of 1,8-cineole affected these transitions in a concentration-dependent manner. At 1% NEO loading, peaks were observed at 80 °C and 260 °C, whereas at 5% NEO loading, the peaks shifted to 73 °C and 256 °C. This shift can be attributed to the plasticizing effect of the oil⁸, which reduced intermolecular interactions and consequently decreased thermal stability. This behavior is consistent with the plasticizing nature of hydrophobic substances, as previously reported for gelatin-based films by Tongnuanchan et al. (2016)^46,47.

Fig. 4.

DSC curves of the G_C1 (black), G_C3 (red), GN1 (blue), GN1_C3 (green), GN5 (purple), and GN5_C3 (yellow).

In addition, a distinct double endothermic peak profile observed in the high-temperature region of crosslinked samples emerged as a characteristic feature. This behavior is attributed to the prolonged exposure of the samples to GTA vapor, leading to a higher effective GTA concentration at the surface and preventing homogeneous completion of the crosslinking reaction throughout the entire bulk. Due to diffusion limitations, particularly in the G_C3, GN1_C3 and GN5_C3 samples, the outer layers were more extensively crosslinked, resulting in a more stable and compact network structure, whereas the inner regions remained either less crosslinked or partially preserved the original gelatin structure due to limited GTA penetration. This structural heterogeneity is reflected in the DSC thermograms as a double endothermic peak. Accordingly, the high-temperature peak corresponds to the outer regions with a higher degree of crosslinking and greater thermal stability, while the lower-temperature peak is associated with the inner gelatin phase exhibiting a lower degree of crosslinking and greater chain mobility⁴⁸.

Water contact angle (WCA) test

The balance between hydrophilic and hydrophobic properties of nanofibers has a very important role as it directly affects cell proliferation and scaffold-tissue integration. Due to the hydrophilic nature of G, the contact angle results of G-based nanofibers generally appear below 90 degrees. As shown in Fig. 5, the WCA of nanofibers crosslinked for 1 h is 43°, whereas for G-based nanofibers crosslinked for 3 h, the WCA increases to 53°.

Fig. 5.

Contact angles of crosslinked nanofibers G_C1 (a), G_C3 (b), GN1_C3 (c), GN5_C3 (d).

Mechanical test

Figure 6 presents the tensile performance of the nanofiber samples. The results show that the neat G nanofibers, without crosslinking and NEO incorporation, exhibited a tensile strength of 5.85 ± 0.35 MPa and an elongation at break of 3.10 ± 0.87%. After crosslinking for 1 h and 3 h, the tensile strength of the G nanofiber membranes changed to 4.95 ± 0.42 MPa and 5.17 ± 0.25 MPa, respectively, while the elongation at break decreased to 2.78 ± 0.32% and increased again to 3.39 ± 0.22%. The incorporation of 1% and 5% (v/v) NEO into the polymer solution led to a decrease in tensile strength. However, the flexibility of nanofibers containing 5% (v/v) NEO was higher compared to those with 1%, suggesting a plasticizing effect of NEO that reduces the intermolecular interactions by creating gaps between polymer chains⁸.

Fig. 6.

Mechanical test results of nanofibers, strain at max. load (a) and tensile strength (b).

In vitro release test

One of the critical parameters for evaluating the functionality of nanofiber membranes is their drug release behavior, which provides essential information for guiding subsequent experimental design⁴⁹. The release profile of NEO from GN1_C3 and GN5_C3 nanofiber mats over 120 and 240 h was investigated (Fig. 7). Since 1,8-cineole constitutes approximately 55% of NEO^6,7 the release profile and loading calculations were based on this major component. This approach enables a more representative evaluation of the primary active compound while maintaining the use of the whole essential oil in the formulation⁵⁰.

Fig. 7.

In vitro drug release profiles of GN1_C3 and GN5_C3 nanofibrous patch: NEO calibration curve (a), release profile from the GN1_C3 nanofibrous patch (b) and GN5_C3 (c). All the measurements were repeated three times, and the errors were less than 5%.

Antibacterial efficiency

The antimicrobial potential of the nanofibers was quantitatively evaluated by disk diffusion test against Gram-positive Staphylococcus aureus and Gram-negative Escherichia coli strains⁵¹(Fig. 8). Gentamicin disk used as a positive control produced large inhibition zones with diameters 20 mm on S. aureus and 19 mm on E. coli, confirming the validity of the test (Table 3). Pure Gelatin (G_C3) nanofibers (Group II) did not show any significant antimicrobial activity against either S. aureus or E. coli; the diameter of 7 mm measured in S. aureus represented only a minimal increase of 1 mm from the diameter of the disk itself (6 mm), while no inhibition zone was observed in E. coli. This finding confirms the biological inertness of the G polymer alone. In contrast, nanofibers containing NEO exhibited concentration-dependent activity against both Gram-positive and Gram-negative pathogens^52,53. While the GN1_C3 group formed inhibition zones with diameters of 8 mm in S. aureus and 7 mm in E. coli, the GN5_C3 group exhibited more pronounced inhibition zones reaching 13 mm in S. aureus and 12 mm in E. coli.

Fig. 8.

Inhibition zones of the produced nanofibers. Positive Group (Gentamycin/10 µg) (I), G_C3 (II), GN1_C3 (III), GN5_C3 (IV), for S. Aureus and E. coli respectively.

Table 3.

Inhibition zone of the added nanofiber diameters in mm.

S. aureus	I	II	III	IV
	20	7	8	13
E. coli	19	-	7	12

Cytocompatibility analysis

Metabolic viability assessment

According to MTT analysis (Fig. 9), 5 × 10³ L929 cells were initially added to each sample, and the results obtained on days 1, 3, and 7 showed that cell proliferation in the Gelatin nanofiber (G_C3) group increased significantly over time. While the viable cell number in the G_C3 group was approximately 2800 on day 1, it reached 9200 on day 7. In contrast, lower metabolic activity was observed initially in the GN1_C3 (1% NEO) and GN5_C3 (5% NEO) groups (~ 1900 and ~ 1000 cells, respectively). However, by day 7, the cell numbers continued to proliferate, reaching around 6300 and 6000, respectively (p < 0.01). This suggests that the addition of NEO initially produced a dose-dependent cytotoxicity, but the cells adapted over time⁵⁴. These results indicate a concentration-dependent reduction in metabolic activity at early time points in NEO-containing groups, followed by a progressive increase over time.

Fig. 9.

MTT assay results of L929 fibroblast cells cultured on gelatin nanofibers (G), and gelatin nanofibers containing 1% (GN1) and 5% (GN5) NEO after 1, 3, and 7 days. Each sample was seeded with 5 × 10³ cells. Error bars represent SD (n = 3). Statistical significance: *p < 0.05, **p < 0.01, ***p < 0.001.

Fluorescence microscopy analysis

The effects of nanofibers on the adhesion and morphology of L929 fibroblast cells were analyzed by DAPI (Blue: Nuclei) and Phalloidin (Green: F-Actin/Cytoskeleton) staining after a 7-day culture period (Fig. 10). Low-magnification (4x) images show high-density and homogeneous spreading of L929 cells in all nanofiber types by day 7, supporting the results obtained from the MTT assay. High-magnification (20x) images revealed that cells adhered to all groups exhibited a typical fibroblast-like, spindle-shaped, healthy morphology⁵⁵. F-actin filaments stained with phalloidin showed well-developed stress fibers and prominent cytoplasmic spreading in both the pure G control group (G_C3) and the NEO containing groups (GN1_C3 and GN5_C3). These observations indicate that all nanofiber groups supported normal cell morphology and cytoskeletal organization under in vitro conditions.

Fig. 10.

Fluorescence microscopy images of L929 fibroblast cells cultured on nanofibers after 7 days, demonstrating cellular adhesion and cytoskeletal organization. Cell nuclei are stained with DAPI (Blue) and the F-Actin cytoskeleton is visualized using Phalloidin (Green). The scale bars are 500 μm for the 4x images and 100 μm for the 20x images.

SEM analysis for cells

The interaction of nanofibers on the surface morphology and spreading of L929 fibroblast cells was examined by SEM on days 3 and 7 (Fig. 11). Images show that the cellular monolayer increased significantly over time in all groups and that the cells adhered to the material surface. By day 3, cells on all three materials had typical fibroblast morphology, with spindle-shaped appearance and apparent spreading efforts on the substrate⁵⁵. At this stage, cell-to-cell contact had already begun. In particular, cell morphology in the NEO containing groups (GN1_C3 and GN5_C3) was similar to the control group (G_C3), demonstrating that the oil did not negatively affect the material’s surface topography or cellular adhesion. By day 7, the surfaces of all materials were completely covered with a continuous cell layer. At this stage, the cells exhibit a flatter and polygonal morphology, establishing tight contacts with neighboring cells and fully spreading. SEM images support that both pure Gelatin and NEO containing nanofibers provide a stable and supportive microenvironment in which L929 cells can grow, and proliferate. These findings indicate that both pure gelatin and NEO-containing nanofibers supported cell attachment and proliferation in vitro⁵³.

Fig. 11.

Scanning Electron Microscopy (SEM) images illustrating the morphology and proliferation of L929 fibroblast cells cultured on the nanofiber materials after 3 and 7 days. The scale bars represent 10 μm (for Day 3) and 20 μm (for Day 7).

Discussion

A decrease in surface tension was observed with the increasing of NEO. At low EO concentrations, the number of molecules migrated to the surface is small, resulting in a minimal effect on surface tension. However, the surface becomes more densely coated, decreasing surface free energy and, consequently, leading to a more pronounced decrease in surface tension at higher EO concentrations. Mohammadi et al. (2022)⁵⁶ loaded coconut EO into PCL/G-based nanofibers and they confirmed that an increase in coconut EO concentration led to a noticeable reduction in surface tension. Similarly, Rafiq et al. (2018) reported that increasing concentrations of clove EO, lavender EO, and cinnamon EO in PVA/alginate-based solutions resulted in a consistent decrease in surface tension⁵⁷. Additionally, Callıoğlu et al. (2020) demonstrated that surface tension decreased with increasing EO concentration in PVP/G-based solutions containing varying concentrations of clove and eucalyptus EOs. It is consistent with trends observed in other polymer–EO systems⁵⁸.

The current results display that the addition of 1% (v/v) NEO to the G solution did not significantly affect electrical conductivity, whereas a higher concentration (5% v/v) resulted in a noticeable decrease. As the amount of EO increases, ion movements are restricted and conductivity decreases. According to the literature, EOs can interact with biopolymer systems and modify solution properties—particularly surface tension and conductivity. For example, in a G system containing Satureja EO⁵⁹, both surface tension and electrical conductivity were reported to decrease significantly. Similarly, the incorporation of ethanolic pomegranate peel extract (EPPE) led to a reduction in electrical conductivity compared to the control G solution²¹. Furthermore, the addition of curcumin to G solutions has also been shown to decrease³⁹. Although studies on the direct effect of NEO on conductivity are not widespread, studies involving its main component, 1,8-cineole, have reported no significant change in the electrical conductivity of PVA/chitosan-based solutions⁶⁰.

According to the results, shear thinning (non-Newtonian, pseudoplastic) behavior was observed in all biopolymer solutions. This indicates that the flow resistance gradually decreases with increasing shear rate. In the GN1 sample, as the shear rate increased, the NEO droplets in the emulsion reduced the internal friction within the system. As a result, the solution flowed more easily and the apparent viscosity decreased. This behavior reflects a typical pseudoplastic property⁶¹.

A number of studies have reported that the addition of EOs to G solutions reduces their viscosity. For instance, Eghbalian et al. (2021) incorporated Mentha spicata EO into a sodium caseinate-G solution, which resulted in a decrease in viscosity and an enhanced in fluidity⁶². Similarly, Mohajeri et al. (2023) fabricated nanofibers by using Plantago psyllium L. seed gum/G solutions containing Cuminum cyminum EO (CCEO). The addition of CCEO to the structure reduced solution viscosity without impairing fiber formation⁶³. However, when the NEO concentration was increased to 5% (v/v), a noticeable increase in viscosity was observed. As the EO concentration increased, the interactions between oil droplets within the solution increased, leading to an increase in viscosity due to the increased heterogeneity of the solution⁶⁴. For instance, Callıoğlu et al. (2020) performed the addition of lavender EO to PVP/G-based solutions and it resulted in increased viscosity while decreasing surface tension⁶⁵. In another study by the same author, the incorporation of clove EO and eucalyptus EO to PVP/G solutions also led to a rise in viscosity⁵⁸. In summary, the NEO used in this study was found to both decrease and partially increase the viscosity depending on its concentration. Furthermore, the viscosity results of G, GN1, and GN5 solutions ensured the production of bead-free, homogeneous, and uniformly dispersed nanofibers. The results were consistent with existing literature.

The morphological properties of nanofibers (diameter, surface roughness, fiber alignment, porosity, etc.) depend on many parameters, with the physicochemical properties of the solution and processing conditions being among the most important. Increasing solution viscosity leads to greater entanglement of polymer chains and increased fiber diameter. Conversely, higher conductivity increases the electrostatic stress on the jet, resulting in finer fibers. Similarly, lower surface tension improves jet formation, resulting in the production of finer, more uniform fibers^30,66.

According to the physicochemical properties shown in Table 1, increasing the EO concentration resulted in higher viscosity and lower conductivity. While this would typically be expected to increase fiber diameter, the observed reduction was attributed to a decrease in surface tension (from 35.09 ± 0.21 to 32.79 ± 0.33 mN/m), which facilitated jet elongation and the formation of thinner, more uniform fibers^67,68. However, 5% (v/v) NEO loaded and 3 h crosslinked nanofibers exhibited the largest average fiber diameter (Fig. 2i). In general, an increase in EO concentration is expected to result in a decrease in the surface tension of the solution, thus producing thinner fibers. However, it was observed that prolonged crosslinking with GTA vapor restricts fiber movement and creates a more compact fiber structure. Consequently, the fibers become dimensionally smaller. After GTA vapor crosslinking, G nanofibers exhibited visible surface yellowing and slight dimensional shrinkage, which can be attributed to fiber compaction and diameter reduction. Similarly, another study comparing one-step and traditional two-step GTA vapor crosslinking methods showed that one-step crosslinking significantly reduced fiber diameters^69,70. Conversely, the prolonged crosslinking time led to an increase in fiber diameter at high EO concentration. A similar trend was reported by Chen et al. (2022), where feather keratin/G nanofibers exposed to extended GTA vapor treatment showed an increase in fiber diameter³³. In another study, gentamicin-loaded PVA/G nanofibers displayed an increase in diameter from 369.26 nm (non-crosslinked) to 447.72 nm after 6 h of crosslinking³². According to these findings, the complex balance between factors such as EO concentration, crosslinking time, and electrospinning parameters needs to be considered and optimized for obtaining the desired fiber structure.

Initially, the WCAs of neat G-based nanofibers crosslinked for 1 and 3 h were evaluated. The results indicated that nanofibers crosslinked for 3 h exhibited an increased WCA. During crosslinking process with GTA vapor, the free amine (–NH₂) and hydroxyl (–OH) groups on the G react with GTA to form covalent bonds, resulting in a decrease in the number of hydrophilic groups on the surface. Consequently, the surface becomes more hydrophobic, leading to an increase in the WCA^18,33. For this reason, all WCA measurements were conducted after applying a 3-hour crosslinking process.

Although the crosslinking process increases the hydrophobicity of nanofibers, the addition of NEO led to a decrease in the WCA. Similarly, a study on the incorporation of NEO into PVA/inulin-based films reported that the WCA increased in some cases and decreased in others. This phenomenon is not solely dependent on the hydrophobic nature of the EO, but also on surface properties (roughness, energy, porosity, and chemical interactions). In the same study, despite the increase in EO concentration in the PIP-N5 sample, the WCA decreased, and it was suggested that this could be due to the higher number of micropores⁸. Similarly, despite the high EO content in the GN5_C3 sample, the observed decrease in WCA may be due to the micropores present in its structure. There are a lot of studies in the literature, the incorporation of EOs leads to a decrease in WCA. For instance, Unalan et al. (2019) performed incorporation of clove EO into PCL/Gel-based nanofibers, and they observed that the WCA decreased as the EO content increased⁷¹. According to the study of incorporation of coconut EO into PCL/G-based nanofibers by Mohammadi et al., a significant decrease in WCA of nanofibers was observed. At lower EO contents (PCL/G/5% (v/v) EO), the oil droplets tended to be embedded within the nanofiber structure, resulting in minimal change in WCA. In contrast, the hydrophobic heads of the oil droplets were reported to be oriented within the matrix, while the hydrophilic heads were oriented outward at higher EO concentrations. Thus, a decrease in WCA occurred due to increased surface hydrophilicity⁵⁶. In another study, Elomar et al. (2024) observed that a decrease in WCA following the sequential addition of chitosan and eucalyptus EO to G-based structures⁴².

Although EOs were incorporated into the G-based nanofibers and different crosslinking durations were applied, the WCA results indicated that the nanofibers retained their hydrophilic character. In tissue engineering applications, the hydrophilic nature of nanofibers is critically important for cell adhesion, biological interactions, and the controlled release of drugs^25,72,73. The fact that the obtained results remained within the hydrophilic range highlights the potential advantages offered by the developed nanofibers.

The observed changes in tensile strength can be attributed to the degree of crosslinking. At 1 h crosslinking time, the decrease in tensile strength may be attributed to the insufficient formation of strong covalent bonds. An increase in crosslinking time to 3 h enhanced tensile strength, indicating improved structural integrity. For the formulation containing 1% (v/v) NEO, the tensile strength increased from 2.41 ± 0.56 MPa (1 h) to 3.70 ± 0.47 MPa (3 h). Similarly, in the formulation with 5% (v/v) NEO, tensile strength increased from 1.29 ± 0.50 MPa (1 h) to 2.27 ± 0.16 MPa (3 h). The results indicate that 3 h of exposure to GTA vapor may be the optimal condition for producing nanofibers with optimal strength and flexibility. However, it should be noted that excessive crosslinking can make the nanofibers more brittle, and the crosslinking process should be performed with due consideration for the nanofibers’ multifaceted functions^32,70,74.

The release profile of NEO from GN1_C3 and GN5_C3 nanofiber mats over 120 and 240 h was investigated. As shown in Fig. 7b, a rapid increase in release was observed during the initial hours for GN1_C3. Approximately 65% of the encapsulated NEO was released within the first 4 h, and the release rate markedly slowed after 24 h. This profile represents a typical high initial burst release. Following this fast release phase, a secondary, slower stage was observed, indicating a biphasic release pattern. By 120 h, approximately 92% of the NEO had been released. Accordingly, out of the 168.12 µg of NEO initially loaded, a total of 155.24 µg was released by 120th hour. A larger fraction of NEO is probably positioned on or near the fiber surface, causing to rapid liberation of most of the content. Kafrani et al. (2018) reported that tannic acid–crosslinked G nanofibers loaded with orange EO exhibited an initial burst release within the first 4 h. They attributed this rapid release to the EO present at or near the surface of the nanofibers, which quickly contacts the release medium and releases it rapidly¹⁴. Similarly, it was suggested that the rapid release in the EO-loaded chitosan/PCL nanofibers developed by Hasanpour et al. (2019) may be due to the oil fraction accumulating on the nanofiber surface⁷⁵.

As shown in Fig. 7c, GN5_C3 (5% NEO) formulation exhibited a much more controlled and sustained release profile compared to GN1_C3. While NEO release was quite limited in the first minutes, it reached approximately 50% by 24 h. After 96 h, the system transitioned to a sustained and slow release phase, reaching approximately 75% by 240 h. The observed slow and sustained release profile is due to the amount of NEO trapped in the internal phase of the nanofiber matrix. Out of an initial 876 µg of NEO loaded into the GN5_C3 nanofibers, a total of 662 µg was released after 240 h. Compared to GN1_C3, GN5_C3 provided a more controlled and prolonged release, making it more suitable for sustained release applications.

In this study, we evaluated the release behavior of NEO as a complex multicomponent system; however, the mechanistic interpretation was primarily based on 1,8-cineole as the major constituent. This approach is widely used in essential oil-based release studies, where a dominant and well-characterized marker compound is employed to represent the overall release behavior of the system^76–78.

The enhanced antimicrobial activity observed in GN5_C3, approximately twice that of GN1_C3, indicates a direct correlation between NEO content and antibacterial efficacy⁵². This broad-spectrum activity is compatible with the mechanism by which NEO the integrity of bacterial cell membranes through its lipophilic components such as terpinen-4-ol^79,80. Inhibition diameters of the GN5_C3 formulation in the 12–13 mm range prove that the material has great antibacterial potential requiring infection control.

Monoterpenes and phenolic compounds, such as 1,8-cineole, α-pinene, and terpinen-4-ol, the main components of NEO, increase membrane permeability by forming lipophilic interactions with the cell membrane, can affect mitochondrial enzyme systems, and can cause oxidative stress-mediated cell damage at high concentrations. Specifically, the suppression of mitochondrial activity by phenolic compounds in essential oils by increasing ROS (reactive oxygen species) production is consistent with the low metabolic activity observed in the MTT test^81,82. However, the hydrophilic and biodegradable structure of the G nanofiber matrix allows for a gradual release of these compounds, allowing cells to regain proliferative capacity between days 3 and 7. Similarly, polymeric nanofibers containing EOs have been reported to support cell proliferation at low doses, while transient cytotoxicity at higher doses, depending on the density of the lipophilic component, has been reported⁸³. These results indicate that the GN1 and GN5 samples initially caused cell stress due to the oil concentration, but over time, the protective effect of the G matrix allowed L929 cells to adapt to the surface and maintain their metabolic activity. Therefore, the developed nanofiber systems demonstrate a cytocompatible profile under in vitro conditions, suggesting their potential for controlled release applications without exceeding cytotoxic thresholds. Importantly, it should be noted that GTA, used as a crosslinking agent in the fabrication process, is known to exhibit cytotoxic effects due to residual aldehyde groups if not completely removed. Residual GTA has been reported to induce cytotoxicity through protein crosslinking and cellular membrane damage, leading to impaired cell viability. Although post-fabrication washing steps were applied in this study to reduce residual GTA, its complete elimination cannot be fully guaranteed. Therefore, the observed cytocompatibility may also be influenced by the efficiency of these removal steps. Future studies should consider alternative crosslinking strategies, such as genipin, enzymatic crosslinking, or physical methods, to further improve the biological safety and clinical translatability of the system⁸⁴.

The observed morphology and cytoskeletal organization indicate that the nanofiber materials provide effective cellular adhesion and are non-cytotoxic. This healthy and orderly structure of the actin cytoskeleton is a key indicator that cells are strongly attached to the substrate surface and are metabolically active. No adverse differences in cellular morphology and cytoskeletal organization were observed between the gelatin (G) control group and the NEO added groups (GN1_C3 and GN5_C3), demonstrating that NEO integrated into the nanofiber matrix did not disrupt cell viability, adhesion mechanisms, or cytoskeletal integrity, despite the mild proliferative suppression seen in MTT results, and that it provided the necessary cellular integration. These findings further support the cytocompatibility of the developed nanofiber systems at the cellular level.

Conclusion

In this study, to minimize the disadvantages of NEO, such as its hydrophobicity, easy evaporation, and unstable structure, NEO was successfully encapsulated within G-based nanofibers and stabilized by crosslinking with GTA vapor for 3 h. Although NEO affected fiber morphology in a concentration-dependent manner, the cross-linking process strengthened the structure and made it suitable for controlled release. GN1_C3 nanofibers completed their release within 120 h, while GN5_C3 fibers, which have a higher oil content, provided extended release up to 240 h due to higher internal retention. Biological evaluations further highlighted the multifunctional potential of the developed system. Disk diffusion assays revealed that GN5_C3 possessed strong antibacterial activity due to its high NEO content and large inhibition zone diameters. Moreover, cytocompatibility studies confirmed that the nanofibers were non-cytotoxic and supported L929 fibroblast adhesion, proliferation, and ECM deposition in a manner comparable to native tissue. Properties such as high mechanical strength, excellent biocompatibility, antimicrobial properties, and controlled, sustained release enhance the potential of these structures as scaffolds for infection control, tissue regeneration, and sustained drug delivery. The stabilization of NEO within the nanofiber matrix further broadens its applicability across diverse biomedical and therapeutic fields.

Materials and methods

Materials

NEO (Melaleuca quinquenervia) was purchased from SNN Natural Products (Delhi, India). According to the product specification, the NEO contained 55% 1,8-cineole and had a density of 0.92 g/mL. Gelatin (Pharmaceutical Grade, Bloom type 240–270) was purchased from Zetamatrix (Istanbul, Turkey) and used as received. Pharmaceutical grade GTA (50% aqueous solution, CAS No. 111-30-8) was purchased from China and used without further purification. Pharma grade Formic acid (CAS No: 64-18-6), acetic acid (CAS No:64-19-7) and ethanol (CAS No: 64-17-5) purchased from Merck (Darmstadt, Germany). The mouse fibroblast cell line L929 (ATCC CCL-1) used in this study was originally obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). As an established commercially available cell line was used, ethical approval was not required. Dulbecco’s Modified Eagle Medium: Low Glucose (DMEM/Low), fetal bovine serum (FBS), and sterile phosphate-buffered saline were all supplied by Biowest. Penicillin/streptomycin was obtained from Capricorn, and trypsin/EDTA solution was purchased from Gibco. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) was sourced from Elabscience. DAPI and Alexa Fluor 488-labeled phalloidin were purchased from Invitrogen. Paraformaldehyde solution, GTA solution, and cacodylate buffer (Thermo Scientific) were utilized.

Methods

Preparation of electrospinning solutions

Pharma grade 18% Gelatin (w/v) was added to pharma grade formic acid: acetic acid solution (70:30) and mixed on a magnetic stirrer for 2 h at room temperature. Then, 1%, 5% and 10% (v/v) NEO was added to gelatin solution and stirred for 2 h at room temperature. It was stirred until a homogeneous mixture was obtained. All optimization results were given in Supporting Information. The solutions and their compositions used in the study are summarized in Table 2.

Table 2.

Overview of the formulations and their contents.

Sample Code	Gelatin Content (wt %)	NEO Content (v/v %)	Duration of GTA vapor treatment	Electrospinning Parameters
G	18	0	Non-crosslinked	Flow rate: 1 mL/h Distance: 10 cm Voltage: 24 kV
GN1	18	1	Non-crosslinked
GN5	18	5	Non-crosslinked
G_C1	18	0	1 h
G_C3	18	0	3 h
GN1_C1	18	1	1 h
GN1_C3	18	1	3 h
GN5_C1	18	5	1 h
GN5_C3	18	5	3 h

Electrospinning process

The electrospinning system consisted of one syringe pump (IME Technologies, Netherland). In order to produce fibers, 5 mL solutions were loaded into syringe and placed on the syringe pump. Optimization studies were conducted at flow rates of 0.2–1 mL/h, a working distance of 10 cm, and high voltage levels between 20 and 24 kV. However, the final electrospinning parameters were set to a working distance of 10 cm, a high voltage of 24 kV, and a flow rate of 1 mL/h. All experiments were conducted ambient temperature at 25 °C, and relatvie humidity 35%. However, the nanofibers loaded with 10% (v/v) NEO did not form a homogeneous morphological structure due to excessive oil loading. Therefore, the nanofibers were fabricated by loading NEO at 1% and 5% (v/v) concentrations in this study. All electrospinning parameters are presented in the Supporting Information (Table SI1–SI10).

Crosslinking process

The crosslinking of the produced nanofibers was carried out using GTA vapor. Aqueous GTA solution (50%) was poured into 35 × 10 mm petri dishes and placed at the bottom of a desiccator. The nanofibers were placed on a glass chamber, and the chamber was incubated under two different temperature conditions (25 °C and 50 °C). Additionally, the nanofibers were exposed to GTA vapor in different volumes (50 µL and 100 µL) and different durations (10 min, 30 min, and 1 h). The crosslinked nanofibers were evaluated in terms of their morphological characteristics and water solubility. The analyses revealed that the most favorable morphology and water resistance were obtained when the nanofibers were exposed to 50 µL of GTA vapor at 25 °C for 1 h. Therefore, in this study, additional analyses were performed by increasing the crosslinking duration from 1 h to 3 h to observe the differences between the 1 h and 3 h crosslinking periods. All crosslinking optimization experiments are summarized in the Supporting Information, based on ESEM analyses and macroscopic images of the nanofiber mats. The corresponding sample codes and details for each condition are presented in Table 2.

Characterization of nanofibers

Physical properties characterization

The viscosity, surface tension and electrical conductivity of the solutions were determined. The surface tension of the solutions was measured with a force tensiometer (Sigma 703D, Attension, Darmstadt, Germany). The viscosity of solutions was determined by a digital a rheometer (Anton Paar MCR302e) at 20 °C. The electrical conductivity was measured using a conductivity probe (Cond 3110 SET 1, WTW, Germany). Each equipment was calibrated before the tests and all samples were tested three times.

Scanning electron microscopy (SEM)

The morphological characterization of nanofibers was also performed with environmental scanning electron microscopy (ESEM-FEI, Quanta FEG 250). The average fiber diameter was determined by analyzing SEM images with ImageJ software (version 1.54 g, National Institutes of Health, Bethesda, MD, USA; https://imagej.nih.gov/ij/ ) and calculating the mean of 100 randomly selected fibers.

Fourier transform infrared spectroscopy (FTIR)

The functional groups of the nanofibers were investigated using a Jasco FT/IR4700 model instrument. The FTIR spectra of samples were founded at the scanning range of 400 and 4000 cm^− 1 and a resolution of 4 cm^− 1.

Tensile testing

Mechanical properties of the both uncrosslinked and crosslinked nanofibers were determined by universal testing device (Shimadzu Corporation, EZ-LX, Kyoto, Japan). The patches were prepared in rectangular dimensions of 10 × 50 mm for testing. Before the test, thickness of the nanofibers was measured using a digital micrometer (Mitutoyo MTI Corp., USA). Tensile test was conducted at a speed of 5 mm/min, applying a 5 kN load cell and was repeated three times to confirm measurement reliability.

Differential scanning calorimetry (DSC)

Thermal properties of the produced nanofibers were determined with DSC (Shimadzu, Japan) with a temperature range of 20–300 °C. The nanofibers were weighed (∼1–3 mg) and put into an aluminum pan. The constant 10 °C/min heating rate was used during the tests.

Water contact angle (WCA) analysis

The nanofibers were cut into a rectangular shape (10 mm × 50 mm) and fixed on glass slides. The water contact angle (WCA) was measured by the sessile drop method using an OCA 15EC tensiometer (Dataphysics Instruments, Germany). A volume of 3 µL distilled water was deposited on the fiber mat surface, and drops were video recorded. The contact angles were calculated from the images at the time of 3 s. The measurements were performed three times for each sample at different locations on the surface.

In vitro release test

The release of NEO from the GN1_C3 and GN5_C3 nanofibers was evaluated over a period of ≈ 5 days for GN1_C3 and ≈ 10 days for GN5_C3. Briefly, 10 mg of NEO-loaded nanofibers were immersed in 3 mL of a release medium consisting of phosphate-buffered saline (PBS, pH 7.4) and absolute ethanol in a 1:1 (v/v) ratio^85–87 and incubated at 37 °C, 100 rpm using a temperature-controlled shaker (TCS10 Thermo Shaker, Ruicheng, China). Samples (1 mL) were collected predetermined time intervals and immediately replaced with an equal volume of fresh PBS: ethanol to maintain sink conditions. The linear calibration curve for NEO was constructed using six concentrations prepared in ethanol, ranging from 7.17 to 459.20 µg/mL (Fig. 7a). Chromatographic detection of NEO was performed at 214 nm using a photodiode array detector (PDA, SPD-20 A) on a reversed-phase high-performance liquid chromatography (RP-HPLC; LC-20 A BLK-Prep, Shimadzu, Japan) system equipped with an Inertsil ODS-3 C18 column (250 × 4.6 mm, 5 μm; GL Sciences, Torrance, USA). The mobile phase consisted of distilled water with 0.1% trifluoroacetic acid (TFA, pump A) and acetonitrile with 0.1% TFA (pump B). Column temperature was performed at 40 °C, with a flow rate of 1 mL/min and an injection volume of 50 µL⁵⁰. Cumulative drug release from the nanofibers was quantified by measuring the absorbance of NEO in the collected buffer solutions and calculating the concentrations based on the calibration curves (n = 3), which were subsequently converted to the released amount (µg) by considering the sampled volume and the total volume of the release medium. NEO calibration curve and cumulative percentage of NEO released over time were calculated by the help of Eq. (1).

Cumulative release (%) X 100 (Eq. 1).

Antibacterial activity measurement

The antimicrobial efficacy of the fabricated nanofiber materials was assessed against two widely recognized reference strains: the Gram-negative Escherichia coli (ATCC 25922) and the Gram-positive Staphylococcus aureus (ATCC 25923), employing the established disk diffusion method. Both bacterial strains were first cultured on Luria Agar (LA) medium at 37 °C overnight. Subsequently, bacterial suspensions were prepared in sterile saline solution and standardized spectrophotometrically to match the turbidity of the 0.5 McFarland standard. This standardized inoculum was then uniformly spread across the entire surface of Mueller-Hinton Agar (MHA) plates using a sterile cotton swab. Disk shape samples of the nanofiber material (6 mm in diameter), were prepared and aseptically positioned onto the inoculated agar surface. Commercially purchased Gentamicin antibiotic disks were concurrently placed on the same plates to serve as the positive control. The plates were then inverted and incubated at 37 °C for 24 h. The presence and diameter of the zone of inhibition surrounding each nanofiber disk and the positive control were measured in millimeters (mm) using a precise measuring tool for determining of quantitatively antimicrobial activity.

Cytocompatibility study

Metabolic viability assessment

Prior to the metabolic viability assessment, the nanofibers (G_C3, GN1_C3 and GN5_C3) were subjected to UV-based surface disinfection before being placed into 96-well culture plates. Both sides of each nanofiber were exposed to UV radiation for 60 min. The sterilized nanofibers were washed with sterile PBS (Biowest) to remove any remaining contaminants. The mouse fibroblast cell line (L929, ATCC CCL-1) was used in this study. Cells were maintained in Dulbecco’s Modified Eagle Medium: Low Glucose (DMEM/Low) (Biowest) supplemented with 10% (v/v) fetal bovine serum (FBS) (Biowest) and 1% penicillin/streptomycin (Capricorn). Cell culture was conducted in a humidified incubator (Thermo Scientific) at 37 °C under a 5% CO₂ atmosphere. Upon reaching confluence, cells were detached from the culture dish using a trypsin/EDTA (Gibco) solution and counted. A suspension of L929 cells was seeded directly onto the surfaces of the sterilized nanofiber materials, which were situated in 96-well plates, at a density of 5 × 10³ cells per well. The nanofibers were incubated at 37 °C in a 5% CO₂ atmosphere for 7 days to facilitate cellular adhesion and proliferation. The metabolic activity of the cells was quantitatively determined on days 1, 3, and 7 using the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) (Elabscience) assay protocol. Following the designated incubation periods, the culture medium in the wells was aspirated and replaced with the MTT solution. The plates were subsequently incubated at 37 °C under a %5 CO₂ for 3 hours. After that, DMSO solution was introduced into each well to fully dissolve the resulting intracellular formazan crystals, and the plates were further incubated for 10 min. Finally, the absorbance values were quantified at 570 nm utilizing a microplate reader (Multiskan SkyHigh, Thermo Scientific). All experimental tests were performed in triplicate.

Fluorescence microscopy analysis

Fluorescence microscopy analysis was performed to examine the adhesion of L929 cells to the nanofibers and their morphological changes. The cells incubated on the nanofibers for 7 days were fixed to visualize the actin cytoskeleton and cell nuclei. For this purpose, the cells were initially rinsed with PBS and subsequently fixed with a 4% paraformaldehyde solution for 15 min at room temperature. Post-fixation, the samples were stained with Alexa Fluor 488-labeled phalloidin (Invitrogen) to delineate cytoskeletal structures and with DAPI (4′,6-diamidino-2-phenylindole; Invitrogen) for the detection of cell nuclei. Fluorescence imaging of the prepared samples was performed using the EVOS fluorescence imaging system (Thermo Scientific).

SEM analysis for cells

SEM analysis was used to meticulously evaluate the morphology and adhesion properties of cells on nanofibers seeded with L929 cells and incubated for 3 and 7 days. After removing the culture medium, the samples were gently rinsed with 0.1 M cacodylate buffer (pH 7.4; Thermo Scientific) and then fixed for 20 min using 2.5% GTA. After fixation, the samples were washed with PBS, dried, and prepared for SEM examination.

Statistical analysis

The statistical analysis was determined by using one-way ANOVA followed by Tukey’s post hoc test. A significance level of p < 0.05 was considered, with data labeled as follows: (*) for p < 0.05, (**) for p < 0.01, (***) for p < 0.001. Fiber diameter measurements were carried out using ImageJ software (version 1.54 g, National Institutes of Health, Bethesda, MD, USA; https://imagej.nih.gov/ij/ ) and calculating the mean of 100 randomly selected fibers.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: Kübra Arancı, Sema Nur Yıldırım; Methodology: Kübra Arancı, Özgür Yılmaz; Formal analysis and investigation: Kübra Arancı, Sema Nur Yıldırım, Elif İlhan, Sümeyra Ayan, Sevgi Gülyüz; Writing - original draft preparation: Kübra Arancı, Sema Nur Yıldırım; Writing - review and editing: Sevgi Gülyüz, Özgür Yılmaz, Oğuzhan Gündüz; Supervision: Sevgi Gülyüz, Özgür Yılmaz. All authors read and approved the final manuscript.

Funding

This study was funded by TÜBİTAK (Science and Technology Research Council of Türkiye) under project number 20AG041. Kübra Arancı is a YOK (Council of Higher Education of Türkiye) 100/2000 doctoral fellow in this research.

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.