2.7.2. Sequential Agar Well Diffusion Test

A variant of the agar well diffusion test was carried out to determine the capacity of the drug-loaded mesh coatings to sequentially create zones of inhibition (ZOIs) over a 14-day period of study. Briefly, several LB agar plates were inoculated with 1 mL of the corresponding bacterial suspension. Then, samples from the different study groups (*n* = 6 per group and strain) were individually placed onto the agar and incubated at 37 ◦C for 24 h. Following incubation, all the plates were photographed for further ZOIs measurement. Those samples exhibiting ZOIs were transferred to freshly inoculated LB plates and incubated again under the same conditions. For each sample, this process was repeated every 24 h until no ZOI was observed in the agar, or up to day 14, if the ZOIs were still being developed. Once the assay was finished, the pictures previously taken were used to quantify the ZOIs amplitude at every 24-h interval. For this quantification, two perpendicular diameters of the ZOIs were measured, using the ImageJ software (National Institutes of Health, Bethesda, MD, USA). Results were plotted in a graph representing the mean ZOI (mm) of the different study groups over time.

### 2.7.3. Bacterial Adhesion to the Meshes

The ability of the different coatings to avoid bacterial adhesion to the mesh surface was assessed by sonication. Samples (*n* = 6 per group per strain) were individually immersed in 3 mL of LB broth, in 6-well plates. Then, each well was inoculated with 1 mL of the corresponding inoculum, and plates were incubated at 37 ◦C for 24 h. Following incubation, the meshes were carefully washed in 1 mL of sterile 0.9% saline to remove the non-adhered bacteria and subsequently transferred to Falcon tubes containing 10 mL of peptone water. The tubes were sonicated for 10 min at 40 KHz in a Bransonic 3800-CPXH ultrasonic bath (Branson Ultrasonics, Danbury, CT, USA). The supernatant in each tube was vortexed for 1 min, serially diluted in peptone water (6 tenfold dilutions), and 100 µL of the supernatant plus serial dilutions were spread on LB agar plates. Following a 24-h incubation at 37 ◦C, the plates were counted to quantify the bacterial adhesion to the surface of the meshes. Results were expressed as the mean bacterial yields per study group.

### 2.7.4. Turbidimetric Determination of the Bacterial Growth

Spectrophotometric assays were carried out to determine whether the presence of the coated meshes inhibited the bacterial growth in culture. As previously described, samples from the different groups (*n* = 6 per group per strain) were immersed in 3 mL of LB broth, inoculated with 1 mL of the corresponding bacteria and incubated 37 ◦C for 24 h. As blank, mesh samples immersed in 3 mL of LB broth plus 1 mL of sterile 0.9% saline were kept under the same conditions. Following incubation, the liquid was aspirated, vortexed, and measured by spectrophotometry (OD600) using an Ultrospec 3100 Pro spectrophotometer (Amersham Biosciences, Little Chalfont, UK). Results were expressed as the mean absorbance recorded for the different study groups.

### *2.8. Statistical Analysis*

The data collected in the different experiments were expressed as the mean ± standard error of the mean. Data were compared by a one-way analysis of variance (ANOVA) and Bonferroni as the post hoc test. All statistical tests were performed using GraphPad Prism 5.0 (GraphPad Software Inc., La Jolla, San Diego, CA, USA). Significance was set at *p* < 0.05.

### **3. Results**

### *3.1. Bioactive Polymeric Film*

The immiscibility of the organic solvent DCM with the diluted acetic acid solution in water provided an excellent tool for the preparation of a biocomposite system, formed by nanoparticles of PLGA loaded with either CHX or RIF dispersed in the diluted aqueous solution of chitosan. The chitosan solution containing the nanoparticles was a homogeneous dispersion of the particles formed after the evaporation of the volatile organic solvent DCM under the stirring of the dispersion of chitosan, thus avoiding the need of isolating such dispersion before the application of the biocomposite system to the mesh materials. Figure 1 shows a micrograph of a dry film of chitosan containing nanoparticles and it is easy to distinguish that the nanoparticles were homogenously and randomly distributed in all the films analyzed by SEM, with a PLGA-CHX nanoparticle size ranging from 300 to 400 nm on average. A similar result with the PLGA-RIF system was obtained. We observed that after the evaporation of the water and sterilization at 40 ◦C, the amount of residual DCM in the dry film was practically zero. The architecture of the different coatings with pure chitosan or chitosan charged with nanoparticles was visualized in SEM after the application to the commercial meshes (Figure 2). The uncoated mesh exhibited its characteristic architecture, consisting of a knitted PP monofilament that created large pores with a honeycomb-like structure. Once coated, the surface of this biomaterial turned different. The polymeric compound formed a thin layer that covered the mesh filaments and the pores in a continuous fashion. *Polymers* **2020**, *12*, x FOR PEER REVIEW 6 of 14 chitosan, thus avoiding the need of isolating such dispersion before the application of the biocomposite system to the mesh materials. Figure 1 shows a micrograph of a dry film of chitosan containing nanoparticles and it is easy to distinguish that the nanoparticles were homogenously and randomly distributed in all the films analyzed by SEM, with a PLGA-CHX nanoparticle size ranging from 300 to 400 nm on average. A similar result with the PLGA-RIF system was obtained. We observed that after the evaporation of the water and sterilization at 40 °C, the amount of residual DCM in the dry film was practically zero. The architecture of the different coatings with pure chitosan or chitosan charged with nanoparticles was visualized in SEM after the application to the commercial meshes (Figure 2). The uncoated mesh exhibited its characteristic architecture, consisting of a knitted PP monofilament that created large pores with a honeycomb-like structure. Once coated, the surface

**Figure 1.** Scanning electron microscopy (SEM) micrograph of PLGA nanoparticles loaded with chlorhexidine, randomly distributed in a chitosan film (scale: 1 µm). The nanoparticles' size ranges from 300 to 400 nm. Similar observations were recorded for the rifampicin-loaded polymer. **Figure 1.** Scanning electron microscopy (SEM) micrograph of PLGA nanoparticles loaded with chlorhexidine, randomly distributed in a chitosan film (scale: 2 µm). The nanoparticles' size ranges from 300 to 400 nm. Similar observations were recorded for the rifampicin-loaded polymer.

**Figure 2.** Macroscopic pictures and SEM micrographs (scales: 200 µm) of (**a**) nude control meshes, (**b**) meshes coated with the unloaded biocomposite (Pol), (**c**) meshes coated with chlorhexidine-loaded biocomposite (Pol-CHX), and (**d**) meshes coated with the rifampicin-loaded biocomposite (Pol-RIF). **Figure 2.** Macroscopic pictures and SEM micrographs (scales: 200 µm) of (**a**) nude control meshes, (**b**) meshes coated with the unloaded biocomposite (Pol), (**c**) meshes coated with chlorhexidine-loaded biocomposite (Pol-CHX), and (**d**) meshes coated with the rifampicin-loaded biocomposite (Pol-RIF).

displayed a more gradual release profile, reaching the total release by day 11. No release was detected

Results from the drug release studies revealed a different response between the two

*3.2. Release Kinetics of CHX and RIF*

### *3.2. Release Kinetics of CHX and RIF*

Results from the drug release studies revealed a different response between the two antimicrobials (Figure 3). The antiseptic drug CHX exhibited a moderate burst release during the first 12 h, followed by a gradual release that was extended for 14 days. However, the antibiotic RIF displayed a more gradual release profile, reaching the total release by day 11. No release was detected in the Pol group, revealing the absence of products of degradation generated by the polymeric coating. *Polymers* **2020**, *12*, x FOR PEER REVIEW 7 of 14 in the Pol group, revealing the absence of products of degradation generated by the polymeric coating.

**Figure 3.** Release profile of CHX and RIF from the chitosan–PLGA nanoparticles polymeric coating over time, following its hydration in PBS (*n* = 3 each). Maximum cumulative release was observed **Figure 3.** Release profile of CHX and RIF from the chitosan–PLGA nanoparticles polymeric coating over time, following its hydration in PBS (*n* = 3 each). Maximum cumulative release was observed after 11 and 14 days for RIF and CHX, respectively.

### after 11 and 14 days for RIF and CHX, respectively. *3.3. Cell Compatibility of the Biocomposite*

a significantly lower percentage viability (*p* < 0.001).

*3.3. Cell Compatibility of the Biocomposite* Cultured fibroblasts exhibited a typical elongated shape and multilayered growth pattern, showing numerous cells undergoing mitosis. This morphology remained unchanged following exposition to the control, Pol and Pol-RIF meshes (Figure 4), keeping fibroblasts their proliferative status. However, cells influenced by Pol-CHX meshes turned stellate and a diminished proportion of mitotic figures was evidenced. Results from the alamarBlue assays revealed equivalent viability among cells cultured in the presence of either the uncoated (99.51% ± 1.10%), Pol (93.21% ± 1.13%), and Pol-RIF meshes (94.65% ± 1.21%), while fibroblasts exposed to Pol-CHX (70.74% ± 2.69%) showed Cultured fibroblasts exhibited a typical elongated shape and multilayered growth pattern, showing numerous cells undergoing mitosis. This morphology remained unchanged following exposition to the control, Pol and Pol-RIF meshes (Figure 4), keeping fibroblasts their proliferative status. However, cells influenced by Pol-CHX meshes turned stellate and a diminished proportion of mitotic figures was evidenced. Results from the alamarBlue assays revealed equivalent viability among cells cultured in the presence of either the uncoated (99.51% ± 1.10%), Pol (93.21% ± 1.13%), and Pol-RIF meshes (94.65% ± 1.21%), while fibroblasts exposed to Pol-CHX (70.74% ± 2.69%) showed a significantly lower percentage viability (*p* < 0.001).

**Figure 4.** Light-microscopy micrographs (scales: 100 µm) of rabbit skin fibroblast cultured in the presence of a (**a**) control, (**b**) Pol, (**c**) Pol-CHX, or (**d**) Pol-RIF material. In each picture, the corresponding mesh is depicted (asterisks). The cells influenced by Pol-CHX coated meshes (c) exhibited slight morphological changes. (**e**) Analysis of cell viability after a 24-h exposure to the coating.

in the Pol group, revealing the absence of products of degradation generated by the polymeric

**Figure 3.** Release profile of CHX and RIF from the chitosan–PLGA nanoparticles polymeric coating over time, following its hydration in PBS (*n* = 3 each). Maximum cumulative release was observed

Cultured fibroblasts exhibited a typical elongated shape and multilayered growth pattern, showing numerous cells undergoing mitosis. This morphology remained unchanged following exposition to the control, Pol and Pol-RIF meshes (Figure 4), keeping fibroblasts their proliferative status. However, cells influenced by Pol-CHX meshes turned stellate and a diminished proportion of mitotic figures was evidenced. Results from the alamarBlue assays revealed equivalent viability among cells cultured in the presence of either the uncoated (99.51% ± 1.10%), Pol (93.21% ± 1.13%),

after 11 and 14 days for RIF and CHX, respectively.

*3.3. Cell Compatibility of the Biocomposite*

**Figure 4.** Light-microscopy micrographs (scales: 100 µm) of rabbit skin fibroblast cultured in the presence of a (**a**) control, (**b**) Pol, (**c**) Pol-CHX, or (**d**) Pol-RIF material. In each picture, the corresponding mesh is depicted (asterisks). The cells influenced by Pol-CHX coated meshes (c) exhibited slight morphological changes. (**e**) Analysis of cell viability after a 24-h exposure to the different materials (*n* = 9 each). Viability of fibroblasts decreased when cultured with Pol-CHX, **Figure 4.** Light-microscopy micrographs (scales: 100 µm) of rabbit skin fibroblast cultured in the presence of a (**a**) control, (**b**) Pol, (**c**) Pol-CHX, or (**d**) Pol-RIF material. In each picture, the corresponding mesh is depicted (asterisks). The cells influenced by Pol-CHX coated meshes (**c**) exhibited slight morphological changes. (**e**) Analysis of cell viability after a 24-h exposure to the different materials (*n* = 9 each). Viability of fibroblasts decreased when cultured with Pol-CHX, compared to the rest of the study groups (#: \*\*\* *p* < 0.001).

### *3.4. Antibacterial Performance of the Biocomposites*

compared to the rest of the study groups (#: \*\*\* *p* < 0.001).

### 3.4.1. Control of the Bacterial Load

The quantification of all the bacterial suspensions used yielded viable CFU values within the expected range. On average, bacterial load from the different inocula were 1.43 × 10<sup>6</sup> CFU/mL and 1.41 × 10<sup>6</sup> CFU/mL for Sa and Se, respectively.

### 3.4.2. Antibacterial Activity over Time

The sequential agar well diffusion test was aimed at determining the ability of the polymeric coatings to exert antibacterial activity over time (Figure 5). As expected, no ZOIs were developed by the control and Pol meshes, while a different response was observed with the antiseptic and the antibiotic-loaded coatings. Regardless of the bacteria inoculated, Pol-CHX meshes developed stable ZOIs that gradually decreased until full depletion following 7 days of the initial inoculation. By contrast, the ZOIs displayed by Pol-RIF were still stable by day 14. The statistical evaluation of the ZOIs amplitude was performed individually for each bacterial strain. When inoculated with Sa, ZOIs from the Pol-RIF meshes were wider than those from the Pol-CHX group, especially at days 1, 2 (*p* < 0.001), 6 (*p* < 0.01), and 7 (*p* < 0.01). Similar findings were collected in the Se-inoculated samples, with statistical relevance at days 1, 2, 3, 6, and 7 (*p* < 0.001).

*3.4. Antibacterial Performance of the Biocomposites*

1.41 × 10<sup>6</sup> CFU/mL for Sa and Se, respectively.

3.4.2. Antibacterial Activity over Time

3.4.1. Control of the Bacterial Load

The quantification of all the bacterial suspensions used yielded viable CFU values within the expected range. On average, bacterial load from the different inocula were 1.43 × 10<sup>6</sup> CFU/mL and

The sequential agar well diffusion test was aimed at determining the ability of the polymeric coatings to exert antibacterial activity over time (Figure 5). As expected, no ZOIs were developed by the control and Pol meshes, while a different response was observed with the antiseptic and the antibiotic-loaded coatings. Regardless of the bacteria inoculated, Pol-CHX meshes developed stable ZOIs that gradually decreased until full depletion following 7 days of the initial inoculation. By contrast, the ZOIs displayed by Pol-RIF were still stable by day 14. The statistical evaluation of the ZOIs amplitude was performed individually for each bacterial strain. When inoculated with Sa, ZOIs from the Pol-RIF meshes were wider than those from the Pol-CHX group, especially at days 1, 2 (*p* <

**Figure 5.** Antibacterial activity over time. Upper panel: Macroscopic pictures of the zones of inhibition (ZOIs) developed by Pol-CHX and Pol-RIF coated meshes at days 1, 3, and 7 of the initial inoculation with (**a**) Sa and (**b**) Se. Lower panel: Representation of the 24-h variation of the ZOIs amplitude of Pol-CHX and Pol-RIF coated meshes challenged with (**c**) Sa and (**d**) Se (\*\* *p* < 0.01; \*\*\* *p* < 0.001). The length of the mesh fragments is depicted (dashed lines). **Figure 5.** Antibacterial activity over time. Upper panel: Macroscopic pictures of the zones of inhibition (ZOIs) developed by Pol-CHX and Pol-RIF coated meshes at days 1, 3, and 7 of the initial inoculation with (**a**) Sa and (**b**) Se. Lower panel: Representation of the 24-h variation of the ZOIs amplitude of Pol-CHX and Pol-RIF coated meshes challenged with (**c**) Sa and (**d**) Se (\*\* *p* < 0.01; \*\*\* *p* < 0.001). The length of the mesh fragments is depicted (dashed lines).

### 3.4.3. Prevention of Bacterial Adhesion

3.4.3. Prevention of Bacterial Adhesion The quantification of bacteria yielded from the different meshes following sonication (Table 1) revealed a strong adhesion to the surface of the control and Pol materials. For both Sa and Se, loads collected from the uncoated meshes were about 2-log smaller than the yields from the Pol materials (*p* < 0.01). Contrary to this, the antimicrobial coatings were highly effective; out of the total Pol-CHX and Pol-RIF meshes inoculated with either Sa or Se, only one sample belonging to the Pol-CHX group exhibited positive Se counts (1.10 × 10<sup>3</sup> CFU), while the surface of the rest of fragments was free of The quantification of bacteria yielded from the different meshes following sonication (Table 1) revealed a strong adhesion to the surface of the control and Pol materials. For both Sa and Se, loads collected from the uncoated meshes were about 2-log smaller than the yields from the Pol materials (*p* < 0.01). Contrary to this, the antimicrobial coatings were highly effective; out of the total Pol-CHX and Pol-RIF meshes inoculated with either Sa or Se, only one sample belonging to the Pol-CHX group exhibited positive Se counts (1.10 × 10<sup>3</sup> CFU), while the surface of the rest of fragments was free of bacteria. Furthermore, the bacterial load found in this sample was significantly lower than the yields from the control and Pol groups (*p* < 0.01).


**Table 1.** Bacterial adhesion to the surface of the different materials. Results were expressed as mean, median, minimum, and maximum bacterial loads (recorded as colony forming units, CFU) yielded per study group and bacteria (*n* = 6 each). The percentage of samples yielding positive counts (%) is provided. from the control and Pol groups (*p* < 0.01). **Table 1.** Bacterial adhesion to the surface of the different materials. Results were expressed as mean, median, minimum, and maximum bacterial loads (recorded as colony forming units, CFU) yielded

bacteria. Furthermore, the bacterial load found in this sample was significantly lower than the yields

### 3.4.4. Inhibition of Bacterial Growth 3.4.4. Inhibition of Bacterial Growth

Bacterial suspensions containing the different meshes were incubated for 24 h to determine whether the antimicrobial biocomposites did influence the growth pattern of Sa and Se, by means of a turbidimetric assay (Figure 6). Data collected from the absorbance recordings revealed a similar response of Sa and Se to the mesh coatings. For both strains, the absorbance slightly increased in those cultures containing Pol meshes compared to the control ones, being this increase statistically relevant for Sa (*p* < 0.05). In turn, and regardless of the bacterial strain, cultures containing Pol-CHX and Pol-RIF exhibited no turbidity and the values of absorbance significantly decreased in comparison with the rest of the groups (*p* < 0.001). Bacterial suspensions containing the different meshes were incubated for 24 h to determine whether the antimicrobial biocomposites did influence the growth pattern of Sa and Se, by means of a turbidimetric assay (Figure 6). Data collected from the absorbance recordings revealed a similar response of Sa and Se to the mesh coatings. For both strains, the absorbance slightly increased in those cultures containing Pol meshes compared to the control ones, being this increase statistically relevant for Sa (*p* < 0.05). In turn, and regardless of the bacterial strain, cultures containing Pol-CHX and Pol-RIF exhibited no turbidity and the values of absorbance significantly decreased in comparison with the rest of the groups (*p* < 0.001).

**Figure 6.** Absorbance (OD600) of (**a**) Sa and (**b**) Se cultures influenced by the different materials (*n* = 6 each). The absorbance of bacteria cultured without mesh is depicted (dashed lines). Regardless of the strain, no growth was recorded when bacteria were exposed to the antimicrobial biocomposites. Statistical significance was as follows: #: vs. Pol (\* *p* < 0.05); ϕ: vs. Pol-CHX (\*\*\* *p* < 0.001); ∆: vs. Pol-RIF (\*\*\* *p* < 0.001). **Figure 6.** Absorbance (OD600) of (**a**) Sa and (**b**) Se cultures influenced by the different materials (*n* = 6 each). The absorbance of bacteria cultured without mesh is depicted (dashed lines). Regardless of the strain, no growth was recorded when bacteria were exposed to the antimicrobial biocomposites. Statistical significance was as follows: #: vs. Pol (\* *p* < 0.05); φ: vs. Pol-CHX (\*\*\* *p* < 0.001); ∆: vs. Pol-RIF (\*\*\* *p* < 0.001).

### **4. Discussion 4. Discussion**

In general surgery, abdominal wall hernia repair constitutes one of the most frequently developed procedures. According to a recent Cochrane review, approximately 20 million surgeries are being performed per year worldwide [23]. In those patients, bacterial mesh infections increase the In general surgery, abdominal wall hernia repair constitutes one of the most frequently developed procedures. According to a recent Cochrane review, approximately 20 million surgeries are being performed per year worldwide [23]. In those patients, bacterial mesh infections increase the morbidity

and mortality rates, lengthen the hospital stay and raise the sanitary costs [24]. Hence, the clinical and social consequences of mesh infections are of major concern.

One of the biggest issues associated with this postoperative complication is that currently there are no gold-standard therapies for preventing mesh-related infections [25]. Preoperative surgical antibiotic prophylaxis is common practice, although it fails to avoid infections in many cases [26] and there is no clear consensus among surgeons regarding its administration [27]. In this milieu, the implantation of antimicrobial meshes constitutes a promising option to reduce the incidence of infection.

Antimicrobial prostheses can be developed following different procedures; the most attractive alternative consists of coating the mesh with bioactive polymeric systems. Ideally, these coatings must cover the surface uniformly, providing an adequate drug concentration at the implant site [28]. Moreover, the coating should be biodegradable, facilitating both infiltration of neoformed connective tissue across the mesh pores and mesh integration into the host tissue [14]. Finally, the drug diffusion from the polymer should be gradual and maintained for about 3–4 days [29,30] to keep bactericidal activity over time. The innovative chitosan–PLGA nanoparticles system designed in this study combines all these requisites, especially when loaded with the antibiotic; firstly, observations from the different experiments carried out revealed signs of polymer degradation, suggesting adequate biodegradability of the biocomposite. Secondly, not only is this coating homogeneously distributed throughout the mesh surface, but also allows for a sustained release of rifampicin for up to 11 days. Nevertheless, the meshes coated with the antiseptic-loaded polymer exhibited an initial burst release during the first 24 h. These findings can be related to the chemical nature of chlorhexidine; this molecule is positively charged and has high binding affinity, features that often hinder the dilution of this antiseptic in polymeric systems and drug carriers [31,32].

Together with the chemical composition, the biocide activity of the selected drug is of great importance when developing antibacterial coatings for biomedical devices. The two antimicrobials included in this study were chosen because of their broad-spectrum activity against Gram-positive and Gram-negative bacteria, as well as their potential usefulness to prevent biomaterial-related infection following hernia repair. Although both drugs are effective biocides, the antiseptic and the antibiotic used have different mechanisms of action. On one side, chlorhexidine interacts with the proteins of the bacterial cell wall, provoking the disruption of the cytoplasmic membrane and the leakage of several cytoplasmic components [33]. By comparison, rifampicin provokes genetic damage caused by a strong, specific inhibition of the bacterial DNA-dependent RNA polymerase that hinders the RNA transcription processes [34]. Observations from our study revealed that polymeric films containing these drugs killed bacteria during the first 24 h of exposure, suggesting a fast, almost immediate biocide performance of these biocomposites, regardless of their mode of action.

Given that mesh infection can arise days, weeks, or even months after surgery [25], the development of antibacterial coatings exerting sustained biocide activity over time would be highly advantageous. It has been suggested that implanted antimicrobial devices should keep activity for 2–3 weeks [30]. Consistent with this statement, our rifampicin-loaded biocomposite exhibits biocide activity for at least 2 weeks. Notwithstanding this, the activity of the coating containing chlorhexidine lasted for at about 1 week. Likewise, rifampicin-loaded materials fully inhibited bacterial adhesion to the mesh surface, while the chlorhexidine-loaded ones allowed mild adhesion of *S. epidermidis* in just one sample. Together, these results reveal a stronger effect and better long-term activity of the antibiotic compared to the antiseptic bioactive system. To the best of our knowledge, no previous data were reported comparing the effect of these two drugs relative to hernia mesh materials, except for another study recently carried out by us [35]. In alignment with this, a clinical trial on the use of antibacterial central venous catheters to reduce bloodstream infections revealed better performance of those devices containing rifampicin compared to others provided with chlorhexidine [36].

Biocompatibility of the polymeric system is another key feature to be considered when designing antimicrobial coatings. In a setting of infection, mesh integration can be hampered if bacteria win the so-called "race for the surface" and colonize the implant before the host cells [37]. Biocompatible devices enhance the interaction of host cells with the implant promoting tissue integration [30,38], thereby indirectly reducing the establishment of infections. Recently, PLGA nanoparticles modified with chitosan were reported as effective drug carriers given their great biocompatibility and sustained release [39]. Our findings align with these observations and show that the biocomposites, especially those provided with rifampicin, exert no detrimental effects on cultured cells.

Given that rifampicin specifically targets the bacterial DNA-dependent RNA polymerase, we confirmed that, at the concentration tested, this drug does not provoke any side effect on fibroblasts in terms of cell viability. However, cytotoxicity of chlorhexidine to eukaryotic cells is well known [33,40]. It was demonstrated that this antiseptic could reduce the growth pattern of cultured cells, as well as provoke morphological modifications on fibroblasts [41], which corroborates our cell viability results. The main reason for this effect is that chlorhexidine can also disrupt the cytoplasmic membrane of eukaryotic cells, even when used at very low concentrations [42]. Together, these findings suggest a potential limitation of this antiseptic for further biomedical applications. In previous studies, our group managed to reduce the toxicity of this antiseptic via incorporation into a quaternary-ammonium-based polymeric compound [17]. With the development of this chitosan–PLGA polymeric system we have optimized the cell response to chlorhexidine, producing a more efficient antimicrobial mesh coating that exerts reduced toxicity while keeping a proper antibacterial activity.

In summary, the results described in the present study demonstrate that a chitosan–PLGA polymeric system can be loaded with antimicrobials to develop a bioactive coating for hernia repair mesh materials. Although the antibiotic rifampicin exhibits an optimal effect, the incorporation of an antiseptic agent such as chlorhexidine also shows an adequate response. Both coatings exert a desirable drug release profile, cell compatibility, or antimicrobial activity. The development of in vivo preclinical studies would provide further data regarding the potential application of these antimicrobial biocomposites to reduce postoperative infections following hernia repair. These steps will be carried out in the next future.

### **5. Conclusions**

Polypropylene mesh materials for hernia repair can be coated with a bioactive chitosan–PLGA nanoparticle polymeric system loaded with either an antibiotic (rifampicin) or an antiseptic (chlorhexidine). Both biocomposite coatings effectively inhibited the bacterial colonization of the mesh without compromising the viability of eukaryotic cells. Rifampicin exerted slightly better performance than chlorhexidine in terms of drug release kinetics, cell compatibility, and antibacterial activity over time. The prophylactic application of these coatings offers a potential strategy to protect meshes from bacterial adhesion following implantation.

**Author Contributions:** Conceptualization, B.P.-K., G.P., J.M.B. and M.R.A.; methodology, B.P.-K., G.P., M.F.-G. and S.B.-M.; software, B.V.-L.; F.G.-M. and L.G.-F.; validation, B.V.-L., G.P., J.M.B. and M.F.-G.; formal analysis, B.P.-K., B.V.-L., M.F.-G. and S.B.-M.; investigation, B.V.-L., F.G.-M. and L.G.-F.; resources, F.G.-M., G.P. and L.G.-F.; data curation, B.P.-K., F.G.-M., M.F.-G. and S.B.-M., writing—original draft preparation, B.P.-K., G.P., J.M.B. and M.F.-G.; writing—review and editing, B.P.K., G.P., L.G.-F. and M.F.-G.; visualization, B.P.-K., L.G.-F., M.F.-G. and S.B.-M.; supervision, G.P., J.M.B. and M.R.A.; project administration, B.P.-K.; G.P., J.M.B. and M.R.A.; funding acquisition, G.P., J.M.B. and M.R.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** The study was supported by Grants SAF2017-89481-P and MAT2017-84277-R from the Spanish Ministry of Science, Innovation and Universities. Financial support from the CIBER-BBN is acknowledged.

**Acknowledgments:** The study was supported by Grants SAF2017-89481-P and MAT2017-84277-R from the Spanish Ministry of Science, Innovation and Universities. Financial support from the CIBER-BBN is acknowledged (CIBER-BBN Internal Collaborations 2018-2021: "Functionalization of polymer coatings using antimicrobial agents for prevention of mesh infection following hernia surgery" and "Polymeric nanoparticles for the treatment of bacterial diseases").

**Conflicts of Interest:** The authors declare no conflict of interest.
