Zinc improves antibacterial, anti-inflammatory and cell motility activity of chitosan for wound healing applications.

We report the successful preparation and characterization of chitosan-Zn complex (ChiZn) in the form of films, intended to enhance the biological performance of chitosan by the presence of Zn as antibacterial agent and biologically active ion. The influence of Zn chelation on morphology and structure of chitosan was assessed by scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy and infrared spectroscopy. The biodegradability study of ChiZn showed a sustained release of Zn up to 2 mg/mL. No toxic response was observed toward stromal cell line ST-2 in indirect contact with the ChiZn films. The dissolution product of ChiZn showed improved wound closure (88% closure) compared to the positive control group (70% closure). Moreover, ChiZn exhibited antibacterial activity against S. aureus together with a slight increase (~30%) in the secretion of VEGF and moderate decrease in nitric oxide evolution. Our findings indicate that ChiZn could be used as a safe and effective wound healing agent.

Introduction

In recent years, a growing deal of attention has been paid to the development of wound dressings from biopolymer-based materials [1]. The repair of a wound is one of the most complex biological processes associated with tissue growth and regeneration controlled by various biochemical and cellular mechanisms. Wound healing is a dynamic process and involves the interaction of a large number of different types of cells [2]. The natural process consists of four overlapping phases: hemostasis, inflammation, proliferation, and tissue remodeling [3]. In the first phase, the clotting mechanism supports the formation of a blood clot to avoid further bleeding [4]. The inflammatory phase occurs immediately after the beginning of hemostasis and lasts for approximately three days. Reactive oxygen species (ROS) are produced by neutrophils and monocytes (which later differentiate to macrophages) in the wound site to prevent infections and enhance the removal of dead cells and remaining debris. The third stage of wound repair is characterized by proliferation and migration of different cell types. Firstly, keratinocytes migrate to the damaged dermis and the fibrin matrix is replaced with fibroblasts to form granulation tissue. In parallel, fibroblasts secrete angiogenic growth factors such as vascular endothelial growth factor (VEGF). During the last phase, granulation tissue converts into a new substrate for keratinocyte migration forming a new and healthy extracellular matrix (ECM) [3], [5]. When the wound cannot repair itself, the healing process needs to be accelerated and stimulated by using various kinds of dressing.

An ideal dressing material should maintain a moist environment and act as a barrier to microorganisms. Wound dressing materials should also be non-toxic, non-allergenic, non-adherent, and easily removable, protect the wound from bacterial infection and possess antimicrobial and anti-inflammatory properties while enhancing the wound healing process [6], [7]. Among biocompatible polymers, a great interest has been given to chitosan due to its favorable properties such as low toxicity and its antioxidant, hemostatic, and immune-stimulating activities [8].

Chitosan is a natural cationic biopolymer which consists of N-acetyl D-glucosamine and D-glucosamine units. Chitosan has a similar structure to glycosaminoglycans which are present in the ECM and are responsible for modulating macrophages, fibroblasts, and endothelial cells by promoting the formation of granulating tissue [6]. Many studies have reported on the use of chitosan as a wound healing accelerator, and in fact, there is evidence that chitosan can be beneficial for every separate phase of wound healing. During the hemostatic phase chitosan can bind to red blood cells which allows rapid blood clotting [9]. It has been also proposed that the interaction between positively charged chitosan molecules and negatively charged bacterial cell membranes leads to the disruption of membrane integrity [9]. Although chitosan shows antibacterial effect in a broad spectrum of microorganisms, its activity is limited to acidic conditions: above pH 6.5 it loses its cationic nature and thus antibacterial activity [10]. Chitosan can promote formation and organization of granulation tissue by modulating the function of inflammatory cells such as macrophages, fibroblasts, as well as keratinocytes and endothelial cells. Chitosan with a relatively high deacetylation degree (89 %) was reported to strongly stimulate fibroblast proliferation [11].

Due to the mentioned properties, chitosan appears highly attractive for wound healing applications. In fact, chitosan has been already exploited on a commercial basis as wound dressing [12]. Moreover, due to its functional groups, chitosan can be functionalized by the addition of metallic ions to enhance its wound healing activity and to improve antibacterial effect. Functionalization of chitosan for biomedical use with metallic ions such as copper [13], [14], [15], [16] and gallium [17] has been reported. Chitosan-metal complexes with specific concentrations of metallic ions were found to have antibacterial properties without triggering a cytotoxic response [13].

Zinc is an essential micronutrient (trace element) that plays a vital role in the wound healing process [18]. It is a cofactor for many metalloenzymes required for cell proliferation, growth, repair of cell membrane and immune system function. For example, zinc-dependent matrix metalloproteinase, which digests dermal membrane and ECM, allows room for cell growth, migration, and angiogenesis during re-epithelization and wound closure [19]. Zinc also can induce hemostasis by affecting clotting factors, platelet aggregation, and interaction with endothelial cells, [20] and can promote clot stability by accelerating clot formation [21]. Besides, it inhibits the growth of several bacterial species [22] and alters early inflammatory responses [23]. The skin contains a relatively high amount of zinc, mainly associated with the epidermis so its deficiency leads to roughened skin and impaired wound healing [19].The complexation of chitosan with zinc ions for different applications such as antimicrobial agents [24], [25], [26], [27], catalysts [28], and water treatment [29] has been reported in the literature [26], [27], [30], [31], [32], [33]. For example, Wang et al. [26] produced chitosan complexes with bivalent metal ions including zinc. The complexes were characterized by XRD, FTIR and elemental analysis indicating that crystalline and structural properties of complexes were different from undoped chitosan: the functional groups of chitosan were considered as the dominating reactive sites. X-ray diffraction measurements revealed the formation of a new crystalline phase in the chitosan-based complex [24], [30]. In vitro antimicrobial activity of the complexes against Gram-positive and Gram-negative bacterial strains, and fungi was much higher than in undoped chitosan or respective metal salts [26]. Moreover, complexation of zinc with chitosan via oxygen and/or nitrogen bindings are likely to leave free donor atoms which could improve the biological activity of chitosan [24]. Therefore, Zn doping potentially enhances therapeutic properties of chitosan, which will be advantageous also for tissue engineering applications. The chitosan-Zn complexes were produced previously for different purposes [24], [27], [28], [29], but very little attention has been paid to the role of chitosan-Zn complexes in wound healing applications so far.

The aim of this study was to produce chitosan-Zn complexes intended for wound healing applications, which should exhibit compatibility with skin-related cells. We report the cytotoxic behavior of the prepared complexes in indirect contact with the stromal cell line ST-2 and long-term cell viability (up to 7 days) in direct contact with fibroblasts. The effect of zinc release on angiogenic properties and migration of keratinocytes was also assessed. The antibacterial activity of the complexes was studied in direct contact with Gram-negative and Gram-positive bacteria strains, and the anti-inflammatory response was evaluated via the determination of nitric oxide release from macrophages. The findings of this study provide a broad perspective for utilization of chitosan-Zn complexes in wound healing applications.

Experimental procedure

Materials

Chitosan (43,000, HQG 10) was purchased from Primex Biochemicals AS (ChitoClear®, Iceland). According to the provider's specifications, the degree of deacetylation is >95 % (determined by colloidal titration method) and the viscosity is lower than 20 cP (1% solution in 1% acetic acid measured by a Brookfield DV-II+ viscometer at 25 °C, and 30 rpm). The average molecular weight of 145 kDa was previously determined by gel filtration chromatography by Pietgat et al. [34]. Glacial acetic acid (>99.85%VWR, Germany), NaOH (VWR, Germany), glycerol (Sigma-Aldrich, Germany), and zinc chloride (ZnCl2, BioReagent, Sigma Aldrich, Germany) were used to prepare chitosan-Zn complex films. Tris (Tris base, Sigma-Aldrich, Germany), HCl (ACS, 37%, Sigma-Aldrich, Germany), acetic acid (99%, Penta s.r.o., Czech Republic), sodium acetate (99%, Sigma-Aldrich, Germany) and lysozyme (hen egg-white, Sigma-Aldrich, Germany) were used to investigate degradation behavior of the films. Lysogeny broth (LB, Luria/Miller) was used for antibacterial studies. Roswell park memorial institute medium (RPMI, Gibco, Germany), Dulbecco's modified eagle medium (DMEM, Gibco, Germany), fetal bovine serum (FBS, Corning, USA), penicillin-streptomycin (PS, Gibco, Germany), phosphate buffered saline (PBS, Gibco, Germany), trypsin-EDTA (0.25%, Gibco, Germany), cell counting kit-8 (CCK-8, Sigma-Aldrich, Germany), rhodamine phalloidin (RP, R415, molecular probes, Thermo Fisher Scientific, Germany), 4′,6-diamidino-2-phenylindole (DAPI, Thermo Fisher Scientific, Germany), Griess reagent (G7921, Thermo Fisher, Germany) and VEGF ELISA (RayBio, Enzyme-Linked Immunosorbent Assay) were used for biological studies. All reagents were of analytical grade and used without any further purification.

Fabrication of chitosan-Zn complex films

Chitosan-Zn complexes were synthesized via the in-situ precipitation method which had been reported previously [13]. Chitosan was dissolved in diluted acetic acid (2% w/v) at a concentration of 2% v/v at 25 °C. After complete dissolution, pre-determined amounts of zinc chloride were added into the chitosan solution and stirred continuously for 2 h to disperse homogeneously. The prepared compositions are summarized in Table 1. The amount of metal salt (m) in the solution is calculated as follows:

Table 1

Quantities of zinc addition to chitosan and corresponding sample labelling.

Sample label	X (%)	ZnCl2 amount (mg)
Chi	–	–
6ChiZn	6	44.9
12ChiZn	12	89.9

where the degree of deacetylation (DDA) used for the calculation is 95 %, mChi is the amount of chitosan dissolved in the solution in g, MMZnCl2 is the molecular weight of zinc chloride 136.286 g/mol), MMGlu and MMN-acetyglu the molecular weight of forming units of chitosan; glucosamine (179.17 g/mol) and N-acetylglucosamine moieties (221.21 g/mol), respectively, and X is the desired ratio of Zn ions related to the content of free amino groups (see Table 1).

The homogenous solution was added into a 0.1 M NaOH solution dropwise. The formed Zn containing chitosan particles suspension was continuously stirred for 2 h at 25 °C. The particles were then separated and washed with deionized water repeatedly until neutral pH. Then, the particles were dried in an oven at 60 °C for 1 day.

To fabricate films, the produced 0.5 wt% of chitosan-Zn complexes and undoped chitosan was dissolved in 1 vol% acetic acid. Then, 10 vol% of glycerol with respect to the total volume of the solution was added as a plasticizer under continuous stirring at 25 °C. The prepared solution was then centrifuged to remove air bubbles and non-dissolved chitosan particles from the solution. Finally, the solution was cast into polystyrene Petri dishes and allowed to dry under a fume hood at 25 °C. After 3 days, the dried films were peeled off and immersed in deionized water. The resulting films weighed 0.38 (±0.04) g and had a thickness of 0.16 (±0.05) mm.

Characterization methods

Scanning electron microscope (SEM) (Auriga Base, Carl Zeiss Microscopy GmbH, Jena, Germany) combined with energy-dispersive X-ray spectroscopy (EDX) (LEO 435VP, Carl Zeiss, Jena, Germany) was used to confirm the presence of zinc in the samples. An electron accelerating voltage of 30 kV was used. FTIR spectra were obtained using an IRAffinity-1S spectrometer (Shimadzu, Japan), with 40 spectral scans in the range 400–4000 cm−1 with the resolution of 4 cm-1.in absorbance mode.

Ion release study

Tris/HCl buffer (Tris base, Sigma Aldrich/ACS, 37%, Sigma Aldrich, 50 mM, pH 7.4) and acetate/acetic acid buffer (99 %, Sigma Aldrich/99% Penta s.r.o., Czech, 0.1 M, pH 4.5) containing 1.5 μg/mL lysozyme (hen egg-white, Sigma Aldrich) was used to determine the in vitro degradation behavior of complex films. The concentration of human serum was considered to be equivalent of lysozyme concentration in degradation buffer solutions [35], [36]. The films of known dry weights (W0) were fixed with cell crowns to immerse them into the solution (1.5 mg/mL) and were incubated in the lysozyme-containing solution in a shaking incubator at 37 °C and 120 rpm. After predetermined time points (4 h, 8 h, 1, 3, and 7 days), the solution was removed and stored for ICP-OES analysis. The samples were rinsed with deionized water, dried at 37 °C for 1 day, and weighed (W1). The weight loss was expressed as a percentage and calculated by the following equation:

To determine zinc release from the films, the method of internal standard was used with an internal standard solution of Sc (10 mg/L) and Be (1 mg/L). At least four calibration solutions were used to create the calibration curve. The experiments were performed in triplicate for each sample.

In-vitro anti-bacterial assay

The antibacterial properties of the chitosan-Zn complexes against Escherichia coli (ATCC 25922, Gram-negative) and Staphylococcus aureus (ATCC 25923, Gram-positive) were determined using the colony forming unit test [17]. Both bacteria strains were cultivated in a lysogeny broth medium on Luria-Bertani agar plates. UV light was used to disinfect the samples for 1 h. The bacteria were grown in lysogeny broth (LB, Luria/Miller) medium at 37 °C for 24 h. Subsequently, the bacterial optical density (OD) was arranged to 0.015 at 600 nm using a spectrophotometer (Thermo Scientific GENESYS 30, Germany). 30 μL of diluted bacterial suspension (~1*107colony forming units (CFU)/mL) were dropped on the surface of the samples and incubated for 6 and 24 h at 37 °C. Then, 3 mL of fresh medium was added and shaken for 30 s to detach the bacteria from the surface of the sample. Subsequently, 30 μL of liquid from detached bacterial suspension were evenly spread onto agar plates. These agar plates were incubated for 24 h at 37 °C to visualize the bacterial growth. The test was performed in triplicate.

Cell culture

Murine stromal cell line (ST-2, Leibniz-Institute DSMZ-German Collection of Microorganisms and Cell Cultures GmnH, Germany), mouse embryonic fibroblast cell line (MEF, ATCC, SCRS 1040), human keratinocyte-like cell line (HaCaT), and murine macrophage cell line (RAW 264.7) were used in the study. RPMI medium was used to culture the ST-2 cells and DMEM medium was used to culture the MEF, RAW 264.7 and HaCaT cells. The mentioned cell culture media were supplemented with 10% Fetal Bovine Serum and 1% Penicillin-Streptomycin. ST-2, MEF and HaCat cells were harvested using trypsin-EDTA. RAW 264.7 cells were harvested using a cell scraper and resuspended in the cell culture medium. In this study, indirect in vitro cytotoxicity test and VEGF secretion assay were performed using ST-2 cells. A direct in-vitro cytotoxicity test was performed using MEF cells and an in-vitro wound scratch assay was performed using HaCaT cells. An anti-inflammatory assay test was performed by using RAW 264.7 cells.

Indirect in-vitro cytotoxicity study

The biocompatibility of the developed materials was determined by using ST-2 cells. Prior to the experiment, all samples and the scaffold holders (Cell Crown 24, Scaffdex, Sigma Aldrich, Germany) were disinfected under UV light for 1 h. Then the samples (films) were inserted into cell crowns and cell culture medium (25 mg/mL) was introduced to each well containing the sample. After 24 h of incubation, the extract of the samples was centrifuged to remove possible particles in the solution. In parallel, ST-2 cell suspension of 105 cell/mL was seeded on the 24-well plates. The medium was changed with the elution extract of the samples and the cells were allowed to grow for 2 days at 37 °C in a humidified incubator with 5% CO2. The regular cell culture medium (CCM) was used as a positive control. After culturing the cells for 48 h, they were washed with PBS to remove the unattached cells and incubated with 1% CCK-8 reagent for 3 h. After transferring the solution to a 96-well plate, the absorbance was measured at 450 nm using a microplate reader (FLUOstar Omega, BMG Labtech, Germany). The cell viability was calculated according to the equation:

Fluorescence microscopy assay

Cell morphology was assessed by staining with rhodamine phalloidin (RP) and DAPI. The cells were fixed using paraformaldehyde concentration of 4 % wt./v in PBS, and immersed in a permeabilization buffer containing Triton X-100. An 8 μL mL−1 RP solution and 1 μL mL−1 DAPI solution were used for staining. Stained cells were observed using of fluorescence microscopy (Axio Observer D1, Carl Zeiss Microimaging GmbH, Germany).

In-vitro VEGF secretion assay

The amount of VEGF released into the culture medium by ST-2 cells after 2 days incubation (indirect test) was measured by a RayBio VEGF ELISA kit. This colorimetric assay is specially designed for detection of mouse VEGF in cell culture supernatants by determination of antibodies specific for the growth factor. The assay was performed according to the manual provided by the supplier.

Direct in-vitro cytotoxicity study

In direct test, 25,000 MEF cells per well were seeded on the top of the samples after sterilization of samples and cell crowns, and the cells were allowed to grow on the films. After the first day, the samples were transferred to the new well plate to remove the cells that have grown on the wall of the wells. The cell viability was measured after 1 and 7 days of incubation.

In-vitro wound scratch assay

An in vitro wound scratch test was performed using human keratinocyte-like HaCaT cells according to a protocol published previously [37]. The extract of the samples with the same concentration (25 mg/mL) was prepared as described above. HaCaT cells were seeded at 500,000 cells per well in a 24 well plate until the cells created a monolayer. After 24 h of incubation, a scratch was created manually by pipet tip. The cells were washed with PBS to remove any debris. Then, the elution extracts were added on the top of the cells. The cell culture medium without any eluates was used as a positive control. During the incubation period, the cell migration was observed after 0, 24, 48, and 72 h by using a light microscope (Primo Vet, Carl Zeiss). The test was performed in triplicate and the width of the scratch was determined using ImageJ software [38].

Anti-inflammatory study

To determine the anti-inflammatory activity of chitosan-Zn complexes, RAW 264.7 cells were used, and the nitric oxide production assay was carried out, as described in literature [39]. The cells were treated with LPS to mimic the inflammation. Then, 50,000 cells per well were seeded in each well of 48-well plates and were allowed to attach for 24 h. Subsequently, 1 μg/mL LPS-containing DMEM cell culture medium with 1% FBS was used to culture the cells for 12 h. Then, the cell culture medium was exchanged by extract of the samples (25 mg/mL). After 24 h, the cell culture medium was collected and Griess reagent kit was used to detect the NO amount in the medium.

Statistics

The differences in analytical parameters between the different samples were analyzed by one-way analysis of variance (ANOVA). The significance level was set at p < 0.05. For comparison of the mean values, the Tukey post hoc test was used.

Results

Chemical and structural characterization

The microstructure, surface characteristics and distribution of zinc in chitosan-Zn complexes were examined by SEM and EDX. An SEM micrograph, elemental maps, and EDX spectra of the produced chitosan-Zn complex films are shown in Fig. 1. The surface of the films was smooth, dense, continuous, without cracks or macropores (Fig. 1-a). The elemental mapping across the sample demonstrated homogenous distribution of zinc (Fig. 1-b). The EDX spectrum of fabricated complexes (Fig. 1-c) revealed the three constitutional elements of chitosan, C (56.56 wt%), O (31.57 wt%), and N (11.56 wt%), due to the presence of N-acetyl-d-glucosamine and D-glucosamine. Low-intensity peaks at 1.012, 8.637 and 9.57 keV were attributed to the presence of zinc (~ 0.30 wt%).

Fig. 1

SEM micrograph (a), elemental maps (b), and EDX results (c) of 6ChiZn complex film.

The structure of Chi, 6ChiZn, and 12ChiZn was evaluated by FTIR spectroscopy. As shown in Fig. 2, chitosan presented a characteristic broad absorbance in the region of 3000–3600 cm−1 because of the O—H and N—H bond stretching [40], [41]. With the incorporation of zinc the frequency of the broad band was shifted to a lower frequency. The band at 1650 cm−1 was attributed to CO (amide I) of N-acetyl unit of chitosan [40] and the band at 1550 cm−1 to bending of N—H (amide II) [40]. The intensity of peaks at 1650 and 1550 cm−1 decreased slightly with an increase of the Zn content in fabricated complexes due to the formation of coordination bonds between -NH2 and Zn. The peak at about 1405 cm−1 is usually assigned to the bending vibration of C—H [42], [43] which gradually disappears in the complex formation. The strongest peak was observed at 1028 cm−1 which is one of the characteristic peaks of chitosan attributed to glycosidic bond (C—O) of glucosamine unit [40]: with increasing Zn content this peak gradually merged into a broad dispersed band, indicating that lone pairs of electrons of oxygen in chitosan interacted with Zn.

Fig. 2

FTIR spectra of Chitosan-Zn complexes with different concentrations of Zn.

Ion release and degradation study

The time dependence of weight loss and release of zinc from fabricated complexes in Tris and acetate buffer containing lysozyme are respectively shown in Fig. 3. The samples in both buffer solutions degraded significantly within the first 24 h. In Tris buffer, the weight loss for 6ChiZn achieved 20 ± 2 % within the first 24 h. Then the degradation continued slowly, and the weight loss reached 23 ± 2% after 7 days of incubation. At the same time the 12ChiZn showed higher degradation rate with 34 ± 1 % weight loss in 24 h and 36 ± 5 % in 7 days. This difference was even higher in the acetate buffer: 6ChiZn showed similar weight loss (21.1 ± 0.1 %) after 24 h. After 7 days the weight loss reached 29 ± 1 %. In comparison, the weight loss of 12ChiZn was 38 ± 3 % and 42.7 ± 0.5 % after 24 h and 7 days respectively. The ion release study showed a trend similar to the weight loss: zinc release was higher in acetate buffer than in Tris buffer solution. The detected zinc release in Tris buffer reached 0.5 ± 0.3 mg/mL and 0.9 ± 0.2 mg/mL within 24 h for 6ChiZn and 12ChiZn, respectively, and did not change significantly up to 7 days. The zinc release was higher in Acetate buffer: 1.3 ± 0.3 mg/mL and 2.1 ± 0.3 mg/mL of zinc was released from 6ChiZn and 12ChiZn within 24 h and a steady release was observed up to 168 h incubation for both complexes.

Fig. 3

Weight loss (%) and zinc release as a function of time in Tris (a and c) and b) acetate buffer (b and d) with lysozyme for different ChiZn samples.

Anti-bacterial activity

The antibacterial activity against S. aureus (Gram-positive) and E. coli (Gram-negative) of chitosan-Zn complexes was evaluated and compared with that of undoped chitosan films to assess their suitability for wound therapy via the colony forming unit test. Table 2 summarizes the antibacterial activity of Chi, 6ChiZn, and 12ChiZn after 6 and 24 h of incubation. The photographs of agar plates after 24 h of incubation are shown in Fig. 4.

Table 2

Antibacterial activity of Chitosan-Zn complex films with different concentrations of Zn compared to neat chitosan against E. coli and S. aureus after 6 and 24 h of incubation. The bacteria growth was classified as +: <10, ++: <150, +++: >300 colony.

Pathogen	Contact time	Chi	6ChiZn	12ChiZn
S. aureus (Gram-positive)	6 h	+ + +	+ +	+
	24 h	+ + +	+ +	+
E. coli(Gram-negative)	6 h	+ + +	+ +	+ +
	24 h	+ + +	+ + +	+ + +

Fig. 4

Images of agar plates after the colony forming unit test performed on Chi, 6ChiZn, and 12ChiZn after 24 h of incubation against S. aureus (Gram-positive) and E. coli (Gram-negative). Scale bar: 1 cm.

The inhibitory effect of Zn-containing complexes was time-dependent: the increase of direct contact time reduced the survival of the bacterial colonies between 6 h and 24 h. However, the incorporation of zinc did not affect the antibacterial properties of the films against E. coli. Although undoped chitosan films did not show any considerable antibacterial activity against S. aureus, the complexes showed significant inhibition effect, eliminating almost all bacterial colonies after 24 h of direct contact with the 12ChiZn complex.

Indirect cytotoxicity with ST-2 cells

To evaluate the biocompatibility of the chitosan complexes, the indirect cytotoxicity test was carried out using ST-2 cells. The viability of ST-2 cells after 48 h of incubation is shown in Fig. 5a. The results revealed that the fabricated complexes have no apparent toxic effect after 48 h of incubation. On the contrary, there was no significant difference between the positive control group (regular cell culture medium indicated as CNT in Fig. 5) and chitosan-Zn complexes. On the contrary, the cell viability on chitosan-Zn complexes was higher than that on the undoped chitosan films, which is already considered a biocompatible material.

Fig. 5

Viability (a), fluorescence microscope images of RP/DAPI stained (b) stromal cells (ST-2) (n = 3,*: p < 0.05, NS: p > 0.05) and vascular endothelial growth factor (VEGF) secretion (c) from ST-2 cells cultured with indicated ChiZn complexes and regular cell culture medium (indicated CNT) for 48 h (n = 9,*: p < 0.05, NS: p > 0.05).

To visualize the morphology of the ST-2 cells, RP and DAPI staining was also performed after the indirect cytotoxicity test (Fig. 5b). The cell viability results coincide with the fluorescence microscopy observations. The density of the cells which are in contact with the extract from undoped chitosan films and from the complexes is similar to that of the control group. No changes in ST-2 cell morphology were observed in the complexes with varying zinc concentrations, and the cells showed their typical morphology for all prepared samples.

Fig. 5c shows the total VEGF secretion after indirect contact with the ST-2 cells after 48 h incubation. The total VEGF release was concentration dependent. 6ChiZn composition secreted similar VEGF amount as the CNT group, which represents the cells without treatment. The12ChiZn composition showed significantly higher secretion of VEGF compared to the CNT group.

Direct cell studies with MEF cells

The viability of MEF cells in direct contact with chitosan films after 1 and 7 days of incubation is shown in Fig. 6. When the cells were in direct contact with the complexes, the viability of MEF cells was significantly reduced with increasing zinc content in the complex. Whereas the absorbance value of 0.33 was reached for 6ChiZn after 7 days of incubation, further addition of zinc decreased the absorbance from 0.16 to 0.14 after 1- and 7-days incubation, respectively.

Fig. 6

Viability of MEF cells in direct contact with the indicated chitosan films after 1- and 7-days incubation (n = 3,*: p < 0.05).

In-vitro scratch assay

The in-vitro wound scratch assay [37] was conducted to compare the migration rate of HaCaT cells of chitosan complexes with that of undoped chitosan. Fig. 7a summarizes % migration of HaCaT cells after 0, 1, 2, and 3 days of incubation with dissolution products of chitosan films. The light microscopy images of the scratches after 0 and 3 days are shown in Fig. 7-b. Compared to undoped chitosan, the fabricated chitosan-Zn complexes displayed superior cell migration. Although % closure was similar with the Control and 6ChiZn until 2 days incubation, the 6ChiZn composition showed the highest migration on day 3 with 88 % of wound closure, while the closure was 70 % and 62 % for the control group and undoped chitosan at the same time point, respectively. In addition, the closure of the wound in the case of 12ChiZn was also observed, although the monolayer of cells was disrupted, and the cell morphology changed to more round-shaped. The undoped chitosan showed a perfect monolayer structure with a slow migration rate.

Fig. 7

% wound closure of HaCaT cells (a) exposed to dissolution products of different Chitosan-Zn films and regular cell culture medium (indicated CNT) after 0, 1, 2 and 3 days and (b) light microscopy images of the scratch of the mentioned samples after 0 and 3 days. Scale bar: 200 μm.

Anti-inflammatory activity

Possible anti-inflammatory potential of chitosan complexes was studied with RAW 264.7 macrophage cells following a previously published protocol [39]. The cytotoxicity of fabricated complexes toward macrophages was evaluated with the WST-8 results after 2 days of incubation and the results are shown in Fig. 8-a The cell viability did not change with the composition. There were no significant differences in cell viability between the positive control and tested specimens, irrespective of the composition. These results can be considered as the indication that the fabricated complexes are not toxic for the used macrophages.

Fig. 8

(a) Viability of RAW 264.7 macrophage cells, (b) % nitric oxide secretion from RAW 264.7 macrophages and (c) fluorescence microscope images of RP/DAPI stained of RAW 264.7 exposed to the dissolution products of chitosan-Zn complexes and regular cell culture medium (indicated CNT) after 48 h incubation (n = 3,*: p < 0.05, NS: p > 0.05).

Furthermore, the macrophages were treated with LPS to create an inflammatory environment, and to transform the macrophages to M1 type (pro-inflammatory) [44]. Then, nitric oxide (NO) release from the macrophages was measured (Fig. 8-b). The LPS induction was documented by a significant increase of NO production of LPS-CNT compared to the CNT group. Moreover, chitosan increased NO production significantly more than LPS-CNT. NO production starts to decrease with the zinc complexation, and the composition 12ChiZn showed significantly lower NO production than undoped chitosan.

In Fig. 8c, the LPS-induced macrophages can be compared with their non-induced counterparts: the RP/DAPI staining images revealed changes of the cell morphology from small and round-shaped to larger in size and irregularly, spindle-like shaped.

Discussion

Chitosan‑zinc complexes were produced by modification of the previously reported in-situ precipitation method [13]. Two different chitosan‑zinc complexes were produced by incorporating Zn2+ ions with the saturation of the free amino groups at 6 % and 12 % (see Table 1), and were denoted as 6ChiZn and 12ChiZn, respectively. According to the SEM micrographs, the fabricated complex films were homogenous and smooth. Similar results were reported previously: chitosan‑gallium complex coatings were also characterized by smooth surfaces [17]. However, when a higher amount of gallium was introduced to the coatings, multiple nano cracks were observed [17]. This phenomenon was attributed to the reduced number of intermolecular and intramolecular hydrogen bonds in the polymer due to the complexation [17]. The examination of the Zn-doped chitosan by FTIR confirmed the formation of Zn-containing chitosan complexes. In the samples with higher amount of incorporated zinc (80 % saturation), the localized presence of ZnCl2 residues in the chitosan matrix was observed by SEM/EDX. The unwanted contaminations were removed by subsequent washing of the samples: the EDX analysis did not detect any Cl residual in the prepared materials after washing. The elemental mapping analyses confirmed that zinc was homogeneously dispersed in the chitosan matrix (Fig. 1-b). The structural changes that occurred in the chitosan matrix with zinc complexation were studied by FTIR (Fig. 2). The shifts and changes in relative absorbance in the FTIR spectra support the assumption that free -OH and -NH2 groups were involved in the complexation process. This assumption is supported by XPS analysis of Cu complexation, as reported by other authors [40]. These results are also in agreement with data reported in the literature previously [13], [41]. Complexes formation between chitosan and metal ions is explained by the Lewis acid-base theory: metallic ions act as an acid which accepts a pair of electrons yielded by chitosan [25]. Although there is no universal agreement about the exact mechanism of metal adsorption process, two models have been commonly suggested to explain the uptake mechanism: the bridge model and pendant model [32], [45], [46] (Fig. 9). In the bridge model, metal ions are bound with several amine groups from the same or adjacent chains [47] (Fig. 9-b), while in the pedant model the metal ions are bound as one to one in pendant fashion to an amine group [48] (Fig. 9-a). According to other reports, both these models most probably coexist. For example, Yazdani et al. [29] prepared chitosan-Zn complexes as a bio-sorbent for phosphate removal from solutions. They reported that -NH2 and -OH functional groups are involved and the molecular structure can be explained by both models [29]. In the present study, a weakened intensity and broadening of the peak at 1100 cm−1 supports the bridge model hypothesis.

Fig. 9

Schematic illustration of the assumed coordination of zinc and chitosan: a) the pendant model and b) the bridge model. The involvement of water molecules and hydrogen bonding has been excluded for the sake of clarity.

An ideal wound dressing material needs to be biodegradable in a controlled manner for adequate cell growth and proliferation without causing any deficiency in the healing process. Several reports have shown that the dissolution of chitosan is controlled by enzymatic degradation, and not by simple dissolution in aqueous medium [49]. Therefore, the weight loss and zinc release in this study were assessed in the presence of lysozyme synthesized from macrophages and neutrophils, which play an important role in the healing process [50]. It has been reported that macrophage-like cells localize near chitosan fibers implanted in a rat model, and play an important role in the degradation of chitosan [51]. It is also known that lysozyme is mainly responsible for the depolymerization of chitosan in the human body [52]. Lysozyme degrades chitosan via hydrolyzing two forming units: the break of β-(1–4) glycosidic bonds occurs, and the initial degradation rate mainly depends on the fraction of N-acetylated units of chitosan [53], [54]. The pattern of degradation of chitosan complex films in this work can be thus explained by an enzymatic degradation process (Fig. 3). Firstly, the broken chains are held in the bulk film until the break reaches the level that lower molecular weight fragments release into the degradation solution [53]. In this study, degradation of complexes reached approximately 35 % in Tris buffer. In the literature, undoped chitosan fibers with similar DDA (93.4 %) showed complete degradation within 5 days of incubation with gradual increase of degradation in phosphate buffer (pH = 7.4) in the presence of 4 mg/mL of lysozyme [51]. The reason for the different results observed in our work could be explained by different concentration of lysozyme and losing of its enzymatic activity with time. A steady trend of sample degradation in the current study supports the idea that appropriate hexasaccharide sequences are lost during the degradation process [49] since the hexameric binding sides are responsible for lysozyme's activity [54], [55]. Similar degradation behavior has been reported for chitosan-Ga complex coatings in PBS-lysozyme solution [17].

Another significant aspect studied in the current work was the degradation behavior of fabricated chitosan complexes as a function of pH. Healthy skin has a slightly acidic pH ranging from 4.0 to 6.0 to keep the skin's barrier function, whereas in wounded skin acidic pH shifts to more neutral pH (7.4). Although the slightly acidic pH is favorable to wound healing by controlling infection and enzyme activity [56], the optimal pH for the proliferation of keratinocytes and fibroblasts is between pH 7.2 and 8.3. Both types of cells rely on the function of zinc-dependent enzymes that play a vital role in the migration of cells [57]. The other aspect that needs to be considered is the composition of the degradation medium: zinc could form insoluble phosphate compounds (such as Zn3(PO4)2, Ksp = 9.0 × 10−33) which may prevent the detection of zinc released to the solution [58]. Thus, to detect and assess the zinc release, the degradation experiments were performed in Tris buffer (pH 7.4) and acetate buffer (pH 4.5), comparatively. A higher release of zinc ions was observed in 12ChiZn due to the overall higher zinc concentration in the complex. For both Zn-containing compositions the zinc concentration was higher in acetate buffer. This data are in line with previous observations, which showed five times higher initial degradation rate of undoped chitosan at pH 4.5 than at pH 7 [59]. Moreover, fluctuation of zinc release from chitosan was observed. A possible explanation for these results is that zinc is leached out from the chitosan surface layer. Thus, over time zinc release slows down due to the longer diffusion paths. Furthermore, the zinc release could increase only after the zinc-depleted chitosan layer is dissolved in the solution and a new surface exposed.

Soft tissue infections occur commonly, therefore the development of wound dressing materials with antibacterial properties is of utmost importance [60]. The antibacterial activity of chitosan is attributed to various mechanisms, such as its chelating ability with trace metal elements [61], and its ability to enter microorganism's nuclei thereby interfering with mRNA synthesis [62]. The most common mechanism is attributed to the polycationic nature of chitosan. The cationic sites interact with the outer membrane of the bacteria or anionic part of bacteria such as lipopolysaccharides. This interaction causes damage to their membrane structure, disturbs metabolic function of bacteria and inhibits their growth. On the other hand, chitosan is a macromolecule, and its mobility is limited when compared to metallic ions. The antimicrobial activity of undoped chitosan is thus expected to be lower than the activity of its complexes. The positive charge on functional groups of the complexes will be higher and the interaction with bacterial strains could be much stronger compared to those of pure chitosan [25], [48]. The interactions of complexes with bacteria surfaces can be then attributed to the interaction of the positive charges of the polymer and the negatively charged bacterial membrane. These opposite charges enhance the attraction by creating electrostatic forces resulting in a strong bond. Once zinc is bound to the bacterial membrane, it can change the membrane permeability, break the membrane integrity, and cause a leakage of cytoplasm thereby leading to cell death [63]. It can be thus concluded that the presence of zinc leads to a higher antibacterial activity against Gram-positive bacteria than against Gram-negative strains, which exhibit higher resistance of their hydrophobic cell walls [63]. It has been shown that the complete inhibition concentration of ZnO nanoparticles against E. coli (≥ 3.4 mM) was much higher than against S. aureus (≥ 1.0 mM) [64]. On the other hand, it has been reported that zinc containing complexes showed a wide spectrum of antimicrobial activities regardless of the type of bacterial strains (both Gram-positive and Gram-negative) after 72 h of exposure. This activity was increased by increasing the content of zinc in the complexes [41]. In the current study, a higher inhibitory effect of chitosan-Zn complexes was observed against S. aureus. The results indicate that the antibacterial activity of zinc compounds may depend on the sensitivity of individual microorganisms, and does not only rely on the Gram-nature of the bacteria [63]. Moreover, previous studies indicated that Gram-positive bacteria are the dominant strains in the initial stage of bacterial infections, while Gram-negative bacterial infections develop in the later stages [60]. In addition, a strong antibacterial efficiency has been observed in the case of chitosan-Cu complexes compared to undoped chitosan against E.coli and Staph. carnosus bacteria strains [13]. The antibacterial properties against both Gram-positive and Gram-negative bacterial strains can be interpreted as bacterial interaction of these two ions (Cu and Zn) activating different mechanisms.

To evaluate the biocompatibility of the fabricated complexes, ST-2 and MEF cells were used both in direct and indirect contact with the prepared samples, since ST-2 cells are versatile cells and fibroblasts play an important role during the healing process. The ST-2 cells did not show any cytotoxic response to chitosan-Zn complexes for both compositions when the cells were exposed to the dissolution products of the chitosan films. The morphology of the ST-2 cells did not seem to be affected. However, 12ChiZn composition showed a decrease in viability of MEF cells in direct contact with the composite. It can be thus concluded that the release of zinc did not result in any cytotoxic behavior. However, the surface properties of films could be the reason for the decreased viability of the MEF cells, since it was only observed in direct contact with the specimens. The cell adhesion and proliferation may have been inhibited due to altered surface properties (i.e. topography) of the produced films with increasing zinc amount. In fact, a decrease of zeta potential of the complexes has been reported in chitosan-Cu complexes with increasing Cu concentration up to an specific amount of copper addition (6 % Cu) [36]. This behavior was attributed to the formation of a polydentate with amine groups of chitosan, which caused depletion of protonated sites in chitosan [36]. Further investigations are necessary to understand the exact underlying mechanisms in the present chitosan-Zn complexes. Interestingly, preconditioned chitosan-Cu complexes showed cytotoxic behavior toward MEF cells in indirect contact due to the high release of copper [13]. On the other hand, chitosan-Ga complex coatings showed no toxicity in direct contact with MG-63 cells [17].

Angiogenesis and formation of granulated tissue are significant events during the last phases of wound healing. The formation of a vascular network is necessary for the conversion of fibrin matrix to granulated tissue. Detection of secreted VEGF from many types of cells including macrophages and fibroblasts is one of the common approaches used for estimating the angiogenic potential of materials since VEGF acts as a biochemical signal to stimulate endothelial cells and keratinocytes to form blood vessels and induce re-epithelization. The produced complexes were therefore used to evaluate the effect of zinc release on VEGF secretion from ST-2 cells. The presence of zinc improved VEGF secretion significantly compared to the control group (regular cell culture medium). The results obtained in this study thus confirm the already published data: even if the VEGF expression was not statistically different on the day 7 in an in vivo test, a diabetic group treated with zinc-vanillin exhibited higher VEGF expression on day 14 compared to the control group [18].

In injured skin, keratinocytes migrate from the wound edge into the defected area through the newly formed dermal scaffolding. Keratinocytes start to proliferate and differentiate to close the wound area and restore the healthy ECM. Therefore, the simple wound healing (scratch) test was performed to evaluate the effect of zinc containing complexes prepared in the present study. 6ChiZn stimulated keratinocytes to migrate without inducing any cytotoxic response in ST-2 and MEF cell lines and exhibited highly enhanced antibacterial activity. These results are in line with those of Tenaud et al. [65] who also observed that the presence of zinc promotes keratinocyte migration. The keratinocytes in the presence of zinc migrated to fill only 8 (±6) % of the scratch area after 6 h. The migration was completed after 24 h, while the positive control achieved only 61 (±9) %. The significant effect of zinc was also observed using HaCaT cell line in modified Boyden chamber: HaCaT cells migrated faster in the presence of zinc compared to the control group [65].

There is considerable evidence that macrophages are a key regulator of the wound healing process. It is well known that the phenotype of macrophages transforms according to the stages of wound healing [44], [66]. Firstly, pro-inflammatory macrophages, referred to as M1, infiltrate the injury site to clean the site from bacteria and dead cells. After tissue begins to repair itself, M1 macrophages transform to anti-inflammatory macrophages referred to as M2 to restore the tissue integrity [44]. The M1 cells produce nitric oxide (NO), which regulates collagen formation, cell proliferation, and wound contraction [67]. M1 macrophages are commonly classified as distended cells with irregular shape and disturbed F-actin in the cytoplasm, whereas M2 macrophages are less lamellar and flattened and enlarged in shape [39], [66]. Several reports have shown that chitosan has a stimulatory effect on macrophages: this effect has been observed by measuring the NO secretion [68], [69]. However, the secretion of NO by chitosan-activated macrophages can also have side effects related to biocompatibility. NO secretion enhances the inflammatory reaction, which also contributes to the destruction of both the polymer and the surrounding tissue, leading to a delay in tissue regeneration [68]. Secretion of NO from RAW 264.7 cells was measured in the present study. The NO release increased significantly compared with the control group. On the other hand, the zinc amount in the samples affected the NO production: an increased amount of zinc reduced NO production. Therefore, zinc containing chitosan could reduce inflammatory response via suppressing NO production induced by LPS. These findings are supported by other studies in this area, describing the presence of zinc and its anti-inflammatory properties [70], [71]. Further investigations such as detection of immune system markers and expression of genes in the presence of different Zn concentration from chitosan-Zn complexes are required to understand the role of zinc on the inflammatory response in the context of wound healing.

Conclusions

Chitosan-Zn complexes with different amounts of zinc addition were prepared by an in-situ precipitation method. The structural features of the films were examined by FTIR and EDX analysis and compared to undoped chitosan. The FTIR spectra confirmed that functional groups of chitosan were involved in the chelation process. A homogenous distribution of zinc was observed by EDX analysis. The cytotoxicity assay confirmed the non-toxic behavior of the synthesized complexes with respect to ST-2 cells. However, the specimen with the highest zinc content showed toxic response in direct contact with MEF cells. Among the developed complexes, 6ChiZn (intermediate) composition demonstrated improved cell viability, cell migration ability, and angiogenetic behavior for effective wound healing and presented antibacterial activity against Gram-positive bacteria strain. The results suggest that the developed chitosan-Zn complexes are suitable candidates for wound therapy, and thus further investigations including in vivo studies in skin wound models should be carried out.

CRediT authorship contribution statement

Nurshen Mutlu: Conceptualization, Methodology, Validation, Investigation, Data curation, Visualization, Writing - Original Draft. Liliana Liverani: Validation, Investigation, Writing – Review & Editing. Fatih Kurtuldu: Validation, Investigation, Writing – Review & Editing. Dušan Galusek: Conceptualization, Writing – Review & Editing, Supervision, Project administration, Funding acquisition. Aldo R. Boccaccini: Conceptualization, Writing – Review & Editing, Supervision, Project administration, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.