Melissa officinalis essential oil loaded polycaprolactone membranes: evaluation of antimicrobial activities and cytocompatibility for tissue engineering applications

PCL (Mn 80 000) was purchased from Sigma Aldrich (Germany). Tetrahydrofuran (THF), was purchased from Merck (Germany). MEO with 99% purity was purchased from local supplier (Biotama, Turkey). Nutrient agar (NA), nutrient broth (NB), potato dextrose agar (PDA), and potato dextrose broth (PDB) were purchased from Sigma Aldrich. Antimicrobial susceptibility testing discs (Oflaxacin, Nystatin, and blank) were purchased from Oxoid (ThermoFisher, USA). The 2,2-diphenyl-1-picrylhydrazyl (DPPH) was purchased from Sigma Aldrich. Cell culture chemicals were purchased from Biowest (France) unless it was stated. The 4',6-diamidino-2-phenylindole (DAPI), bovine serum albumin (BSA) and Triton X-100 were obtained from Sigma Aldrich. Microbial species used in antimicrobial activity tests, are given in table 1, were obtained from American Type Culture Collection (ATCC). L929 mouse fibroblast cell line was obtained from ATCC.

The antimicrobial activity of the MEO was carried out using disc diffusion assay as described previously [22]. Microbial strains used for antimicrobial activity tests are given in table 1. The 100 μl; 0.5 McFarland bacterial suspension spread on NA and 2 McFarland yeast and fungal suspensions spread on PDA medium. The blank discs were impregnated with 20 μl of MEO and placed on the inoculated agar. For bacteria ofloxacin (5 μg/disc), for yeast and fungus nystatin (100 unit/disc) discs were used as positive control (PC). The inoculated plates were incubated at 36 ± 1 °C for 24 h for bacteria and 48 h for yeast, at 27 ± 1 °C for 72 h for fungus. The antimicrobial activity was expressed as the inhibition diameters (mm). Experiments were performed in triplicate.

PCL membranes loaded with MEO were prepared by solvent casting method. The 10% PCL (w/v) was dissolved in THF and stirred at room temperature (RT) overnight for complete dissolution of the polymer. Then, MEO was added to the homogeneous PCL solution at different weight ratios as given in table 2 and stirred overnight at RT. The homogenous PCL, MEO mixture was poured into glass Petri dishes and the solvent was evaporated under fume hood. The Petri dishes were covered with aluminum foil and small holes opened for controlled drying. The membranes were taken from Petri dishes after 48 h. PCL membrane without MEO was used as a control group.

Antimicrobial activity tests of MEO loaded PCL membranes were conducted according to ISO 22196 with little modifications against selected two gram-negative, one gram-positive bacteria and a fungal specie given in table 1. Membranes were cut into 60 mm diameter to get rid of curly edges, both sides were UV sterilized for 1 h and placed into sterile Petri dishes. The 0.5 McFarland microbial suspensions were diluted to approximately 10⁶ CFU ml⁻¹ in 1/500 NB and 1/500 PDB with sterile distillate water for bacteria and fungus, respectively. The 500 μl of diluted inoculum was poured onto membranes and covered with sterilized polypropylene (PP) films. Microorganisms and membranes were held in direct contact with bacteria for 24 h at (36 ± 1) °C and with fungus for 48 h at (27 ± 1) °C under relative humidity of above 90%. PCL membranes without MEO were used as negative control. Microorganisms were recovered from membranes by washing with 10 ml soya casein digest lecithin polysorbate broth (SCDLP) which was prepared as described in the standard. Inoculum containing SCDLP medium was diluted by ten-fold serial dilutions in PBS and inoculated onto relevant agar medium. After incubation at conditions as given above colonies were counted.

The antimicrobial activity was calculated using the following equation:

$$\begin{equation}R = \left( {{U_t} - {U_0}} \right) - \left( {{A_t} - {U_0}} \right) = {U_t} - {A_t}\end{equation} \tag{ 1 }$$

where;

Ris the antimicrobial activity;

U₀is the average of the common logarithm of the number of viable cells recovered from the untreated test specimens immediately after inoculation;

U_t is the average of the common logarithm of the number of viable cells recovered from the untreated specimens after 24 and 48 h for bacteria and fungi, respectively.

A_t is the average of the common logarithm of the number of viable cells recovered from the treated test specimens after 24 and 48 h for bacteria and fungi, respectively.

Surface morphologies of PCL/MEO membranes were characterized by scanning electron microscopy (SEM, FEI, Quanta 650, USA) analysis. Average pore diameters of PCL/MEO membranes were calculated from SEM images by Image J software (NIH, USA). The presence of specific chemical groups in the membranes were analyzed by Fourier transform infrared spectroscopy (FTIR, PerkinElmer, USA) in the 400–4000 cm⁻¹ wavelength range. Hydrophilicity of the membranes were determined by contact angle measurement (Kruss, Germany) using sessile drop method at the end of the 5 s after contact of a droplet with the membrane. The mechanical properties of the membranes were investigated using tissue analyzer Stable Microsystems (TA.XT Plus, England). Samples were cut into 40 × 10 mm dimensions and the initial distance between the grips was set to 30 mm and the crosshead speed was set at 1 mm s⁻¹ (load cell of 5 kg) under dry conditions. Elongation at break (%), tensile strength (MPa), and Young modulus (MPa) were analyzed.

Antioxidant activities of PCL membranes incorporated with different amounts of MEO were determined by DPPH assay. The 0.3 mM DPPH solution was prepared in methanol. The 0.01 g from each membrane were weighted and placed in 1.5 ml of DPPH solution in Eppendorf tubes and kept for 2 h at RT in the dark, experiments were done in triplicate. The 200 μl of solution from each sample were transferred to 96 well plate and the radical inhibition was determined based on decrease in absorbance of DPPH at 517 nm. The inhibition percentage was calculated by the following formula:

$$\begin{equation}{\text{Inhibition}}\left( \% \right) = \left[ {\left( {{A_{{\text{control}}}} - {A_{{\text{sample}}}}} \right)/{A_{{\text{control}}}}} \right] \times 100\end{equation} \tag{ 2 }$$

where;

A_controlis the absorbance of DPPH solution without membrane,

A_sampleis the absorbance of DPPH solution with membrane.

L929, mouse fibroblast cell line was used in cell culture studies. Cells were cultured in Dulbecco's Modified Eagle Medium High Glucose containing 10% (v/v) fetal bovine serum, 1% (v/v) L-glutamine and 10 units ml⁻¹ penicillin, 10 μg ml⁻¹ streptomycin solution. Membranes were cut into 12 mm diameter and both sides were sterilized under UV light for 1 h. The oil including membranes tended to float. To prevent the floating of the membranes and the attachment of the cells to the culture plate surface, the surface was coated with sterilized 2% (w/v) agar solution. Before cell seeding, membranes were conditioned overnight in cell culture medium. L929 cells were seeded on conditioned membranes at a concentration of 2 × 10⁴ cells/membrane in culture medium. Cell seeded membranes were cultured at 37 °C in 5% CO₂ atmosphere. Cell culture medium was changed on days 2 and 4.

Viability of L929 cells on PCL membranes, were determined by resazurin assay (n = 5). To measure the mitochondrial activity of the cells, 0.8 mM resazurin stock solution (PBS including Ca⁺² and Mg⁺²) was diluted 1:10 in the culture medium. Cells were cultured at 37 °C in 5% CO₂ atmosphere for 3 h in resazurin including culture medium. At the end of the incubation period, 200 μl of used cell culture medium was transferred to a 96 well plate. Absorbance values were recorded at 570 nm with 600 nm reference wavelength (BMG Labtech, Spectrostar Nano, Germany). Membranes which did not include the cells were used as the material control groups. Results are reported as % cell viability relative to the viability of the cells cultured on PCL control membranes.

The morphology of the cells was examined by SEM analysis on days 1 and 6. Culture medium was aspirated and the membranes were washed twice in PBS. The 2.5% (v/v) glutaraldehyde (in PBS) solution was added to the membranes as a fixation agent. They were kept in a fume hood for 20 min at RT. Excess glutaraldehyde solution on the membranes was aspirated. Samples were washed again with PBS twice. Samples were kept in PBS at +4 °C until the day before SEM analysis. The samples were dehydrated by keeping the membranes in an increasing series of ethanol solutions (30%, 50%, 70%, 90% and 100%, v/v) for 2 min and finally treated with hexamethyldisilazane for 5 min. After complete dry gold coated samples were examined by SEM.

Nucleus of the cells cultured on membranes were determined by DAPI staining. The samples were washed twice in PBS on day 6 and fixed as in the SEM analysis described above. Fixed cells were permeabilized in 0.1% (v/v) Triton X-100 containing PBS for 10 min. They were washed with 1% (w/v) BSA containing PBS (PBS/A) for three times. Then, the samples were incubated in DAPI reagent (1:1000) containing PBS/A in the dark at RT for 30 min. After incubation, the samples were washed with PBS/A for three times. The images were taken by inverted phase contrast microscope with a fluorescent attachment (Leica, Dmil Led Fluo, USA).

GrapPad Prism9 software was used statistical analysis. One-way Analysis of Variance (ANOVA) and Tukey–Kramer post hoc tests were used to determine the significant differences among the groups (at least n = 3).

PCL based membranes loaded with MEO were synthesized by solvent casting method described in the previous section. It was observed that the opacity of membranes was directly proportional to the amount of MEO concentration. Membranes synthesized at all MEO concentrations (PCL:0M, PCL:0.25M, PCL:0.5M, PCL:1M) are homogeneous.

The surface properties of the PCL/MEO membranes were investigated by SEM analyses. Addition of MEO affected the surface morphology of the membranes as shown in figure 1. The porous structure of the PCL/MEO membranes were increased with increasing concentration of MEO. In addition, the homogeneous pore distribution in all MEO-added membrane groups indicates that the MEO is homogeneously distributed in the polymer structure. Also, it has been revealed that pore size of membranes is directly proportional with MEO amount. Average pore diameter of control PCL membranes without MEO was calculated as 2.5 ± 0.5 μm. Pore diameters of PCL/MEO membranes were increased with increasing MEO content. In the presence of 0.25M, 0.5M and 1M, pore diameters were calculated as 5.1 ± 0.9 μm, 9.4 ± 1.5 μm, and 19.0 ± 7.4 μm, respectively.

Zoom In Zoom Out Reset image size

FTIR spectra of PCL and PCL/MEO membranes are presented in figure 2. Asymmetric and symmetric vibrations of CH₂ group were observed at 2925 and 2855 cm⁻¹, respectively. The peak 1723 cm⁻¹ also shows the C=O vibration of the ester group. The peak at 1166 cm⁻¹ is because of the symmetric COC stretching. The band at 1294 cm⁻¹ is assigned to the backbone C–C and C–O stretching modes in the crystalline PCL [23, 24]. Besides, the peaks 2925 and 2855 cm⁻¹ (CH₂ group) are stronger for PCL:1M membrane than PCL:0M membrane. The peak at 3008 cm⁻¹ refers to the methyl groups exist in phenolic compounds of MEO [25]. This peak was not seen PCL:0M membranes and sharpened with increasing MEO concentration.

Zoom In Zoom Out Reset image size

The wettability properties of the membranes were determined by measuring the water contact angle of the membranes. Evaluation of the hydrophobicity of membranes is important for tissue engineering applications. Some studies in the literature have stated that cell adhesion is higher on hydrophilic surfaces [26]. The contact angle values of the membranes are given comparatively in figure 2. In the case of untreated PCL, the measured contact angle was found about 103.44° and decreased to 83.36° for PCL:0M and PCL:1M, respectively. According to these results; with increasing MEO concentration, the hydrophilicity of membranes was increased.

This increment in the hydrophilicity of the MEO loaded membranes could be related to the penetration of water into membrane's porous structure that leads the water absorption and reduction of contact angle [27]. Previous studies state that cells generally prefer to attach and spread on hydrophilic surfaces, while others can proliferate on hydrophobic surfaces. These results can be related to surface topographies of materials and cell types [28].

Cellular attachment, proliferation, and differentiation can be affected by the mechanical properties of the surfaces [29]. In this study, the mechanical properties of MEO loaded membranes were evaluated in terms of strength, toughness, Young's modulus and elongation at break. The results of the mechanical tests of the membranes are given in table 3. The strength value (MPa) in the PCL control group membrane decreased from 21.517 ± 0.857 to 1.338 ± 0.435 with the increasing MEO ratio. The lowest values were determined in the membrane containing the highest concentration of MEO (PCL:1M).

The mechanical properties of human skin may differ due to the heterogeneity of the skin. The reference values of mechanical properties of human skin given as 2.9–150 MPa for Young's modulus, 1–32 MPa for tensile strength and 17%–207% for elongation at break [30]. Therefore, the mechanical properties of PCL/MEO membranes are comparable to the mechanical properties of human skin. Table 3 shows that PCL/MEO membranes properties matches closely skin mechanical properties.

Figure 3 shows radical scavenging activity of PCL membranes incorporated with different percent of MEO. According to the results, all membrane samples including control PCL membrane showed radical scavenging activity. PCL:0M radical scavenging activity can be correlated to the absorption of DPPH due to the porous surface of the membrane [31]. It has been noted that MEO has the ability to act as a donor for hydrogen atoms or electrons in the conversion of DPPH to its reduced form DPPH-H [32]. According to our results, the increase in the amount of MEO in membranes resulted with a regular increase of free radical scavenging activity. These results are consistent with previous studies that reported the antioxidant activity increased with increasing MEO concentration [8, 33, 34].

Zoom In Zoom Out Reset image size

Antimicrobial activity of MEO against four gram-positive, five gram-negative, one yeast and one fungal species was performed by disc diffusion assay. In the literature, the antibacterial mechanism of action for EOs has generally been expressed as altering bacterial cell permeability by disrupting the structure of the cell membranes. As a result of change in the cell permeability, all other cellular functions are also compromised, including membrane function, efflux pump activity, and respiratory activity [35–37]. It has been proven, using flow cytometry experiments, that EOs alter bacterial cell permeability in the same way in both gram-positive and gram-negative bacteria [38, 39].

MEO shows antioxidant, antihistamine, antispasmodic, antitumor, antiviral and antimicrobial properties due to its wide variety of components. The main components of MEO are citral, citronellal, linalool, geraniol, β-caryophyllene-oxide, tannins, triterpenic acid, flavonoids, terpenes, rosmarinic acid and caffeic acid [40]. Citronella and geraniol are the two compounds that are responsible for disturbing the outer membrane of microorganisms. Therefore, the liposaccharides are easily released from the cell and the permeability of the cytoplasmic membrane to adenosine triphosphate (ATP) is increased. Withdrawal of ATP causes depletion of cellular energy storage and cell death [25].

Antimicrobial activity results of MEO are given in figure 4 and disc diffusion images are given in figure 5. Obtained results revealed that MEO exhibited similar antimicrobial activity against all tested microbial strains. These results are consistent with the study conducted by Mimica-Dukic et al that reported the antimicrobial activity of n-hexane extracted MEO against bacteria, yeast and fungi [8]. MEO revealed similar effect on gram-positive, gram-negative bacteria, yeast and fungus. According to these results, it was predicted that the loading of MEO on the membranes would have a positive effect on gaining antimicrobial activity, especially in membranes used in biological applications.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

In the literature, there are various studies intended to develop antimicrobial biomaterials to be used for different purposes by using MEO. Râpă et al were encapsulated Melissa and Dill EOs into bovine tendon, rabbit skin derived collagen and chitosan mixture by electrospinning to improve the antimicrobial activity for potential wound dressing applications [41]. In another study, MEO loaded methylcellulose hydrogels showed antifungal activity against C. albicans and were proposed as a potential therapeutic for use in the prevention of oral infections [42]. Sani et al studied the antibacterial effect of zinc oxide and MEO loaded chitosan based composite films for food packing applications. They found that the increasing concentration of zinc oxide and MEO increased the antibacterial properties of chitosan films [25].

A quantitative assessment of the surface antimicrobial activity of the PCL/MEO membranes tested against E. coli, S. aureus, P. aeruginosa, and A. niger, according to ISO 22196 with slight modifications and results are summarized in table 4. According to the surface activity results, after 24 h of exposure to the MEO loaded PCL membranes, no viable cell counts of microorganisms were recorded for any sample, except PCL:0.25M. Although 174 CFU cm⁻² S. aureus colonies were counted on NA medium for PCL:0.25M sample, the antibacterial activity still corresponds to a very high value of 99.9%. According to obtained results MEO added PCL membranes, even with the lowest percentage of MEO, showed excellent antimicrobial activity against all tested microorganisms.

3.4.1. Resazurin assay

The viability of L929 mouse fibroblast cell line on PCL/MEO membranes were determined by resazurin assay as given in figure 6. The attached cells on PCL:0M membrane was significantly higher when compared to cells on PCL/MEO membranes (p < 0.001) on day 1. Instead of low cell attachment ratio of MEO loaded PCL membranes all of them supported cell proliferation on day 6 when compared to the viabilities on day 1. Here, cell attachment was initially thought to be influenced by MEO amount rather than hydrophilicity of the membranes, which depends on pore structure. The increasing amount of MEO in the structure of the membranes may inhibit protein attachment which may result with decreased cellular attachment at day 1. The difference between PCL:0M, PLC:0.25M, and PCL:0.5M were decreased up to 10% on day 6. However, L929 cells on PCL:1M membrane showed significantly lowest cellular viability when compared to the viability of the cells on other membranes (p < 0.05, p < 0.01).

Zoom In Zoom Out Reset image size

In the literature, the effects of MEO or extracts of different parts of M. officinalis on various cell lines were investigated. In a study, the effect of MEO and citral (one of the major components of MEO), was investigated on glioblastoma multiforme cell lines. Increased DNA fragmentation and caspase-3 activation of the cells indicated that they both trigger the apoptosis in these cell lines [43]. The cytotoxic activity of MEO and thyme oil loaded nanoemulsions were investigated with Caco-2 cell line (human colon adenocarcinoma). Two different viability assays demonstrated that the MEO loaded nanoemulsions decreased cellular viability more than thyme oil loaded nanoemulsions at the same concentrations [44]. In another study, MEO was applied to HaCaT, human keratinocytes and BEAS-2B, human bronchial epithelial cells at 0.13%–0.001% (v/v) concentrations in culture medium. CC₅₀ (cytotoxic concentration) were calculated by MTT assay and found as 0.0012% and 0.015% (v/v), for HaCaT and BEAS-2B, respectively. High antibacterial activity of MEO was also determined with more than 10 bacterial strains in the same study and it has been suggested that this antibacterial and cytotoxic activities may be due to monoterpene aldehydes in the structure [45].

In contrary, neuroprotective effects of M. officinalis extracts and MEO has been shown in different studies [46–48]. MEO was applied to primary neuron cultures obtained from mouse embryos for 2 h in serum free medium and hypoxia model was triggered at 5% O₂ concentration. Cell viability and cell staining analysis showed that 10 μg ml⁻¹ MEO was protective against in vitro hypoxic conditions [46]. In another study, Hassanzadeh et al indicated that water extract of M. officinalis leaves at 10 μg ml⁻¹ concentration significantly reduced ecstasy induced neuronal death in hippocampal neuronal culture. In addition, increased concentration of MEO to 100 μg ml⁻¹ triggered the death of hippocampal neuronal culture [47]. When the ethanolic extract of M. officinalis on human leukocytes was examined between 10 and 150 μg ml⁻¹ no genotoxic and cytotoxic effects were observed [48]. Our results were also showed that the effect of MEO on L929 cell line was concentration depended in a very narrow range. The proliferation of L929 cells on PCL:0.25M and PCL:0.5M was almost as good as the proliferation of L929 cells on PCL:0M membranes at day 6, although these membranes initially favored cell adhesion only slightly compared to the PCL:0M membranes on day 1.

3.4.2. SEM analysis and DAPI staining

Attachment behavior and morphology of the cells were determined by SEM analysis. Results were compatible with the viability values indicating less cell numbers on PCL/MEO membranes on day 1 when compared to the cell numbers on PCL:0M membranes (embedded small images in figure 7). In addition, the cells showed rounded morphology on MEO containing membranes on day 1. L929 cells were covered all the surface of PCL:0.25M and PCL:0.5M on day 6 and especially they formed second layer on PCL:0.25M. Spreading behavior of some of the L929 cells were clearly seen on PCL:1M, however, they were not able to proliferate as much as the cells on the other membranes. As a result, PCL:1M significantly suppressed the proliferation of L929 cells.

Zoom In Zoom Out Reset image size

The morphology of the cell nucleus on the membranes was determined by DAPI staining on day 7 (figure 8) There was no significant difference in the structure of the nucleus of the cells on the membranes. Distribution of the cells on the membranes was as the same as in the SEM images and was also correlated with the viability results.

Zoom In Zoom Out Reset image size