Introduction

Chitosan is a linear amino polysaccharide. It is the second most abundant natural biopolymer after cellulose. Chitosan is derived from the cell walls of fungi, crustacean shells, exoskeletons of invertebrates and arthropods [1,2,3,4,5]. Chitosan is available in various grades; they differ in the molecular weight and deacetylation degree [6, 7]. Owing to its intrinsic properties such as biocompatibility, biodegradability, and hemostatic, antibacterial, antitumor, and antioxidative activities, chitosan has received increasing attention as a unique biopolymer for widespread applications in drug delivery and tissue engineering [4, 8,9,10,11]. Besides the ability of chitosan in accelerating the wound healing process, its antibacterial and antimicrobial activities against a wide variety of pathogens make it suitable for wound dressing applications [12,13,14,15,16]. Antibacterial properties of chitosan are explained as: the chitosan has a positive surface charge in physiological pH; it effectively binds to the negatively charged membrane of various bacterial cells resulting in the rupture of the cell wall membrane and killing the bacteria [17, 18].

Bioadhesive properties of chitosan prevent fluid or air gap formation lessening the risk of infection and reducing pain and increasing the longevity of the wound dressing at the wound site [19]. In addition, chitosan shows excellent ability in film formation when dissolved in organic acids, which is another interesting feature of this polymer over other biopolymers used in the fabrication of wound dressings [20,21,22]. An ideal wound dressing should be biocompatible and should keep the wound adequately moist. It should also protect the wound against the invasion of pathogens, accelerate the healing process, and should be non-antigenic, hemostatic, and elastic [5, 23, 24]. Chitosan itself can satisfy most of the criteria for an ideal wound dressing material; however, the chitosan membranes have weak mechanical properties, which impede the chitosan in being a suitable material for wound dressing applications. Crosslinking is one of the methods that can improve the mechanical properties of the polymers [25]. Several reagents are available for the crosslinking of chitosan, including glutaraldehyde, ethylene glycol, genipin, tripolyphosphate, diglycidyl ether, diisocyanate, and dicarboxylic acids [26, 27]. Non-cytotoxic crosslinking agents, like most of the dicarboxylic acids, can be used in the crosslinking process without further specific steps to remove the residues of crosslinker. Dicarboxylic acids alone cannot dissolve chitosan, but they can be utilized in crosslinking reaction under certain conditions [27]. This study aimed to develop a membrane for wound dressing application with mechanical properties, cytocompatibility, and stability compared to plain chitosan. Here, we investigated different dicarboxylic acids, non-toxic reagents, to fabricate crosslinked chitosan membranes for wound dressing applications. To evaluate the efficacy of each crosslinker, the membranes were first characterized by ATR-FTIR spectroscopy, weight loss in water, and tensile strength test. We performed biological assessments like cell attachment and growth on the best candidate.

Material and methods

Materials

Chitosan (Mn = 810,000, DD > 85%) was purchased from Sigma-Aldrich (USA). Adipic acid, oxalic acid, sebacic acid, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), dimethyl sulfoxide (DMSO), ethanol, and glutaraldehyde were purchased from Merck (Germany) and used without any purification. RPMI 1640, fetal bovine serum (FBS), penicillin/streptomycin, l-glutamine, trypsin/EDTA, and phosphate buffer saline (PBS) were purchased from the GIBCO Company.

Preparation of crosslinked chitosan films

The chitosan solution was prepared by dissolving the polymer (1% w/v) in aqueous acetic acid solutions (1% v/v) and stirring for 12 h at room temperature. The mixture was filtered using Büchner funnel to separate undissolved particulates. Finally, the solution was mixed with crosslinkers, then cast into Petri dishes, and incubated at 25 ± 2 °C for 24 h. The ratios between the amine groups of chitosan, based on the degree of deacetylation (DD), and carboxyl groups of the acids were used to determine the quantity of dicarboxylic acids. The chitosan films were peeled off from the Petri dishes and were further dried in a vacuum for 12 h. Finally, the chemical crosslinking of the chitosan membrane was carried out by heating the specimens at 90 °C under vacuum for 2 h. The membranes were kept at room temperature in a desiccator without further treatment for the upcoming tests. A plain chitosan film was also prepared through the same procedure and used as control. Figure 1 shows the schematic illustration of the chemical structure of dicarboxylic acids and crosslinking reaction between amine groups of chitosan and carboxyl moieties of dicarboxylic acids.

Fig. 1
figure 1

Chemical schematic of the mechanism of crosslinking of chitosan by dicarboxylic acids

Full size image

ATR-FTIR spectroscopy

The chitosan films were characterized using ATR-FTIR spectroscopy (Bruker, Equinox 55, Germany) with a resolution of 20 scans cm−1, in transmission mode. The samples were analyzed in the wavenumber range of 400–4000 cm−1.

Mechanical properties

The mechanical properties of both crosslinked and non-crosslinked chitosan films were determined according to ASTM D882-12 [28] using a tensile strength instrument, GALDABINI (Sun 2500) with a load cell of 50 N capacity. The films were cut into 50 mm length and 10 mm width and placed in a vacuum oven at 60 ± 5 °C overnight. The specimens were fixed vertically on the grips of the instrument and pulled at a crosshead speed of 1 mm/s. The gauge length between the two clamps was set at 10 mm to determine Young's modulus (E), ultimate tensile strength (UTS), and elongation at break (εb) for each specimen.

Weight loss in an aqueous medium

The chitosan films were cut into square shapes, weighed (w1), immersed in double-distilled water, and incubated at 37 °C (± 2 °C) for 24 h. Next, the specimens were dried in a vacuum oven at 60 °C (± 5 °C) overnight and weighed again (w2). The weight loss value (Ws) was calculated using Eq. (1).

$$ W_{{\text{s}}} \;\left( \% \right) = \frac{{w_{1} - w_{2} }}{{w_{1} }} \times 100 $$
(1)

Cell adhesion study

The cell adhesion study was performed using the L929 cell line (supplied by national cell bank of Iran, Pasteur Institute of Iran) according to the same procedure as described by Jamalpoor et al. [29]. Briefly, the number of 4 × 104 cells was seeded on the chitosan membranes (non-crosslinked and crosslinked), with dimensions of 20 × 20 mm2 and sterilized with UV radiation for 30 min in advance. Cell-seeded films were incubated for 4 h at 37 ± 1 °C, allowing the cells to attach. Then, 400 µL of RPMI-1640 medium supplemented with 10% FBS was added to each well and incubated. To assess the morphology of the cells, we removed the culture medium and rinsed the membranes with the PBS solution. The attached cells were fixed with 4% GTA solution for 20 min and dehydrated in graded ethanol solutions (50%, 60%, 70%, 80%, 90%, and 100%), and finally dried at 25 ± 2 °C, for scanning electron microscopy (SEM) (AIS 2100, Seron Technology, Korea). In this study, tissue culture plate (TCP) was considered as the negative control.

Cytotoxicity

The cytotoxicity of the membranes was studied with the L929 cell line utilizing tetrazolium dye-based colorimetric (MTT) assay. The MTT assay was performed on the extracts of the films, according to Bonakdar et al. [30], with minimal modifications. Briefly, the extract of each sample was prepared according to the ISO 10993-12 standard test method [31] by incubating the specimens in RPMI 1640 culture medium for 24 h. The RPMI 1640 medium free of the sample was also incubated and used as a negative control. The L929 cells were seeded into a 96-well plate at the confluency of 1 × 104 cells per well and supplied by 90 μL RPMI 1640 medium and 10 μL FBS and incubated for 24 h. Then, the medium was removed and replaced with 90 μL of the prepared extract media plus 10 μL FBS and incubated for 24 h. Next, the media was removed, and 100 μL of MTT dye solution (0.5 mg/mL in PBS) was added to each well followed by incubation for 4 h. Finally, the MTT solution was aspirated out, and 100 μL of DMSO was added to each well and incubated for 15 min to dissolve the formazan crystals prior to the absorbance measurement. The optical density (OD) was performed by a microplate reader (ELX800, Biotech, USA) at 570 nm and normalized to the control OD.

Statistical analysis

All the results are expressed as mean ± standard deviations. Statistical comparisons were performed with one-way analysis of variance (ANOVA). The p value < 0.05 was considered as a statistical significance level.

Results and discussion

Characterization

ATR-FTIR spectra of both non-crosslinked and crosslinked chitosan films are shown in Fig. 2. The results are separated in different colors. To avoid overwriting the data of one sample on the others, we introduced gaps between the graphs; consequently, the vertical axis is representative of actual intensities of the peaks. The characteristic transmittance peak around 3450 cm−1 in each spectrum is attributable to the overlapping N–H and O–H stretching bands of chitosan, while the peaks seen at 1565 cm−1 are related to the C=O stretching vibration of the amide I group [32, 33]. Formation of ionic salts of chitosan with carboxylic acids is confirmed by the appearance of the NH3+ peaks around 1514 cm−1 and 1615 cm−1, and –COO− peak at about 1556 cm−1 [34, 35]. Intensities of the peaks of the carbonyl group (1500–1600 cm−1) and amine group (1400 cm−1) are influenced. The observation is due to the interactions of chitosan with the organic acids; the results are in good agreement with the published literature [36]. Also, similar observations have been reported for chitosan/glutamic acid and chitosan/succinic acid blend hydrogels [37]. The peak that appeared around 1300 cm−1 in the spectra of the oxalic acid-crosslinked specimens confirmed the formation of amide II as a result of the reaction of amine groups of the chitosan and carboxyl groups of the oxalic acid at 90 °C under vacuum, confirming the report of Cai et al. [25].

Fig. 2
figure 2

ATR-FTIR spectra of non-crosslinked and crosslinked chitosan films in transmittance mode

Full size image

Mechanical properties

The primary purpose of using dicarboxylic acids for crosslinking of chitosan is to improve the mechanical properties and increase the durability of chitosan films for biomedical applications with high biocompatibility [36]. The results of the mechanical properties of the chitosan films are summarized in Table 1. The oxalic acid reagent enhanced both Young's modulus (from 726.9 to 1042.4 N/mm2) and the UTS (from 62.1 to 75.6 N/mm2) in the chitosan films. Crosslinking with oxalic acid, however, decreased the elongation at the break of the chitosan films from ~ 40 to ~ 15%. Thus, it could be considered as an effective crosslinking agent for chitosan. Crosslinking by the other dicarboxylic acids decreased the mechanical properties of the chitosan films (no statistical significance), which can be attributed to the lower reactivity of the carboxylic moiety as a result of their long carbon backbone. The long carbon chain, in general, leads to lower reactivity of the end chemical groups. As a fact, the mechanical properties (mainly, Young's modulus, UTS) of the crosslinked chitosan membranes will decrease in the wet condition, because water molecules act as a plasticizer in the polymeric matrix. Thus, it is rational to select the membrane with higher mechanical properties (membranes crosslinked with oxalic acid) as the best candidates.

Table 1 Mechanical properties of the non-crosslinked and crosslinked chitosan membranes
Full size table

Weight loss

Chitosan-based biomaterials have poor stability in aqueous medium and thus need further treatments such as crosslinking for biomedical applications [38]. The results of the weight loss tests of the chitosan films are represented in Fig. 3. All the crosslinked chitosan films showed weight loss after 24 h immersion in double-distilled water, but there was significantly less weight loss in crosslinked membranes than plain chitosan films. Chitosan/oxalic acid represented the smallest change in weight. The results suggest that oxalic acid was the most effective crosslinking reagent for chitosan since it led to the lowest Ws compared to the others. According to the results of the mechanical properties and the weight loss percentages of the specimens, the oxalic acid-crosslinked chitosan membranes were selected as the optimized specimen for further studies.

Fig. 3
figure 3

Weight loss percentages of chitosan-based membranes after 24 h

Full size image

Cell adhesion study

An ideal wound dressing material should not exhibit any cytotoxic effect during the time of exposure to the wound. Thus, evaluation of the cytocompatibility of a wound dressing material is critical. In this study, the L929 cell line was selected as the model cell line since the fibroblast cells play crucial roles in the wound healing process [39]. SEM images of the cultured fibroblast cells on the surfaces of the chitosan films are represented in Fig. 4. As seen, the fibroblast cells adhered well to the surface of the films and spread. No visible cell debris or changes in morphology, such as cell necrosis and loss of spindle shape, were observed, which certify the non-cytotoxicity of the prepared films. Proper attachment of the fibroblast cells on the surface was expected based on previous reports [40, 41], and our results delineated the same trend. The morphologies of the cells on the crosslinked membranes with 0.22% w/v oxalic acid appear to be diverted from the spindle shape of the healthy fibroblast. It could be ascribed to the surface properties of the membranes and the residual oxalic acid in the polymer matrix. This observation is in agreement with cytocompatibility results, indicating the optimum amount of the oxalic acid should be 0.11% w/v for the given chitosan substrate.

Fig. 4
figure 4

SEM images of: a chitosan, b chitosan/oxalic acid (0.11% w/v), c chitosan–oxalic acid (0.22% w/v), and d TCP

Full size image

Cytocompatibility

Further confirmation of the non-cytotoxicity of the prepared films was conducted with MTT assay on the extracts of the specimens. Figure 5 demonstrates the viability of the fibroblast cells after 24 h contact with the extract of the samples. According to the results of MTT assay, the cell viabilities of the chitosan and chitosan/oxalic acid (0.11% w/v) films were close to the TCP. They did not show any significant difference (P > 0.05), translating in the nontoxicity of the membranes. The published scientific literatures [16, 42, 43] can support this observation.

Fig. 5
figure 5

Cell viability of the L929 fibroblast cells in the extracts of the chitosan and oxalic acid-crosslinked chitosan membranes

Full size image

Conclusion

In this study, the crosslinking effects of various dicarboxylic acids, including oxalic acid, adipic acid, and sebacic acid, on the properties of chitosan membranes were evaluated. The obtained results showed that the oxalic acid could significantly improve the mechanical properties and stability of the chitosan membranes. The water loss percentage data were a means of confirming the oxalic acid efficacy in the crosslinking of chitosan, which led to the minimum value of 5.4%, translating in a stable crosslinking reagent. Cell viability assay for the oxalic acid-crosslinked chitosan membrane showed the cytocompatibility of the membranes. The findings suggest that oxalic acid (with a concentration of 0.11% w/v) could be a promising crosslinking reagent of chitosan with almost no cytotoxicity for biomedical wound dressing applications.