The role of glycerol in manufacturing freeze-dried chitosan and cellulose foams for mechanically stable scaffolds in skin tissue engineering

International Journal of Biological Macromolecules 275 (2024) 133602 Available online 2 July 2024 0141-8130/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). The role of glycerol in manufacturing freeze-dried chitosan and cellulose foams for mechanically stable scaffolds in skin tissue engineering Katarína Verˇ cim´ akov´ a a , Joanna Karbowniczek b , Marian Sedl´ aˇ r a , Urszula Stachewicz b , * , Lucy Vojtov´ a a , * a Ceitec – Central European Institute of Technology, Brno University of Technology, Purkyˇ nova 656/123, 612 00 Brno, Czech Republic b Faculty of Metals Engineering and Industrial Computer Science, AGH University of Krak´ ow, al. Adama Mickiewicza 30, 30-059 Krak´ ow, Poland ARTICLE INFO Keywords: Chitosan Cellulose Ionic complex Glycerol Tissue engineering ABSTRACT Various strategies have extensively explored enhancing the physical and biological properties of chitosan and cellulose scaffolds for skin tissue engineering. This study presents a straightforward method involving the addition of glycerol into highly porous structures of two polysaccharide complexes: chitosan/carboxymethyl cellulose (Chit/CMC) and chitosan/oxidized cellulose (Chit/OC); during a one-step freeze-drying process. Adding glycerol, especially to Chit/CMC, significantly increased stability, prevented degradation, and improved mechanical strength by nearly 50%. Importantly, after 21 days of incubation in enzymatic medium Chit/CMC scaffold has almost completely decomposed, while foams reinforced with glycerol exhibited only 40% mass loss. It is possible due to differences in multivalent cations and polymer chain contraction, resulting in varied hydrogen bonding and, consequently, distinct physicochemical outcomes. Additionally, the scaffolds with glycerol improved the cellular activities resulting in over 40% higher proliferation of fibroblast after 21 days of incubation. It was achieved by imparting water resistance to the highly absorbent material and aiding in achieving a balance between hydrophilic and hydrophobic properties. This study clearly indicates the possible elimination of additional crosslinkers and multiple fabrication steps that can reduce the cost of scaffold production for skin tissue engineering applications while tailoring mechanical strength and degradation. 1. Introduction Chronic or acute wounds are still one of the largest worldwide healthcare burdens [1]. Additionally, treatment of wound infections remains a great challenge as the healing process is long and requires customized drug delivery [2]. Therefore, the basic science related to the development of wound dressing materials and skin scaffolds allowing to enhance the regulation and recovery process are one of the biomedical science priorities [3]. Huge amounts of studies and preclinical trials have been conducted over the past decades to develop material-assisted skin regeneration strategies to ensure rapid progress of the right sequence in the healing process. Typically, materials are combined with cultured human fibroblasts or keratinocytes, which is associated with high cost, long preparation time 3–4 weeks, poor cell attachment, difficult handling due to thin cellular layers, poor mechanical stability, and antigenic response [4]. Other material modifications include combination with various pharmaceuticals or antibacterial agents such as silver or active chlorine [5], nanoparticles [6] growth factors [7], plant-based oils [8] or natural material like hyaluronic acid [9] to improve bioactive properties and accelerate healing. Bioresorbable materials are highly desired for wound healing applications. Collagen, gelatin, cellulose, alginate, chitosan, silk fibroin, fibrin, keratin, casein, or fucoidan are biomaterials investigated the most frequently in various forms of wound dressings, meshes, hydrogels, 3D composite materials, matrices, and fibers [10]. Fabrication methods such as solvent casting/particulate leaching, freezedrying, gas foaming, electrospinning, 3D bioprinting, micro-patterning and micro-molding have been widely used for the fabrication of bioengineered tissue scaffolds [4]. A frequent drawback in biomaterial development is the poor mechanical properties of the final scaffolds and the need for physical or chemical crosslinking. Typical crosslinking agents are genipin [11], glutaraldehyde [12], tripolyphosphate (TPP) [13], 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC) with N- * Corresponding authors. E-mail addresses: [email protected] (K. Verˇ cim´ akov´ a), [email protected] (J. Karbowniczek), [email protected] (M. Sedl´ aˇ r), [email protected] (U. Stachewicz), [email protected] (L. Vojtov´ a). Contents lists available at ScienceDirect International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac https://doi.org/10.1016/j.ijbiomac.2024.133602 Received 28 March 2024; Received in revised form 21 June 2024; Accepted 30 June 2024 International Journal of Biological Macromolecules 275 (2024) 133602 2 hydroxysuccinimide (NHS) [6], epichlorohydrin [14], citric acid [15], etc. In addition, these biomaterials can face high costs due to the multiple fabrication steps along with the possible toxicity of crosslinkers deteriorating their biocompatibility. Many synthetic polymers have also been studied due to their greater mechanical properties e.g., polyvinyl alcohol, polyethylene glycol, poly[lactic-co-glycolic acid], polycaprolactone, and polylactic acid, reaching the skin values, which varies from 5 kPa to 140 MPa, depending on the location [16]. However, among their major disadvantages, there is a slow biodegradation rate and cytotoxic subproducts during biodegradation, which negatively impact cell interactions and has been widely linked to inflammatory responses. Furthermore, hazardous and expensive organic solvents are often required for their solubility. Some synthetic materials also require chemical or physical crosslinking to meet the criteria of their intended application. In case of skin treatment, they require addressing of strong hydrophobicity, poor swelling index, brittle structure, and lack of cellrecognition signals [17]. For this study, chitosan (Chit) and water-soluble cellulose derivatives have been chosen as they are the most abundant and renewable resources on Earth and have numerous applications in the field of tissue engineering and wound healing. They are usually designed in the form of scaffolds, hydrogels, films, membranes, nanoparticles, fibers, etc. [18]. Chitosan is a positively charged antibacterial biopolymer with a hemostatic effect involving β-(1,4) glycosidic bonds between D-glucosamine and N-acetyl-D-glucosamine, allowing further modification with other materials, resulting in excellent elasticity, flexibility, and a lower inflammatory response [7]. Furthermore, the degradation of chitosan produces harmless amino sugars that can be completely absorbed by the human body. There is always a need to tailor chitosan properties for precise tissue requirements, having predictable pore sizes, degradation rates, improved mechanical properties, and cell-material interaction [19]. Carboxymethyl cellulose (CMC) has a wide range of applications in biomedicine and pharmacy due to its thickening and emulsifying properties. CMC is responsive to ionic strength and pH and possesses compatibility when blended with other polymeric solutions due to its polyelectrolyte nature. This property is crucial for the preparation and study of tissue scaffolds and drug delivery systems [20,21]. CMC is also typically used to absorb wound exudates and autolytic debridement from inflamed and chronic wounds by attaching to microbes and inhibiting microbial growth [22]. It has also been extensively studied to improve hemostatic properties, and its bioresorption is due to a hydrophobic polysaccharide backbone and many hydrophilic carboxylic groups [23]. Oxidized cellulose (OC) is a cellulose derivative with growing potential in soft tissue engineering. It is characterized by nonimmunogenicity, antibacterial and anti-tumor activity, high absorbability, antiadhesive effects, high hemostatic activity, and promising as a carrier for controlled drug delivery [24,25]. The structure of OC presents anhydroglucose units linked by β-d-1,4 glycosidic bonds. Aldehyde and hydroxyl groups of the OC can react with other functional materials and improve the final biological and mechanical properties [26]. The bioresorption of OC is enabled via chemical depolarization and enzymatic hydrolysis mediated by glycosidases, which leads to nontoxic final products of glucuronic acid and glucose [24]. Glycerol has been chosen for this study as a cheap, non-toxic, and well-known biocompatible polyol, typically used in a wide range of industrial applications, including food, cosmetics, pharmaceuticals, biofuels, biosensors, and bio-diagnostics. It is also used as a plasticizer for biopolymers to greatly improve the flexibility and workability of rigid materials by decreasing intermolecular interactions [27]. Plasticizers can intercalate and break interactions of 3D networks in which polymer chains are tied by secondary forces by establishing bonds with the polymer and masking the secondary bonds between the macromolecules [28]. The intermolecular interactions between chitosan and cellulosebased on long-range attractive forces of oppositely charged polymer chains [29,30], ensure suitable materials for the encapsulation and controlled release of bioactive molecules [31,32]. Chitosan solubilization occurs in acidic media and the presence of primary amines along the backbone of the chitosan structure forms polycation as a result of protonation of NH 2 [33]. CMC is an anionic polyelectrolyte prepared by swelling cellulose in an aqueous NaOH solution, followed by carboxymethylation with sodium salt of monochloroacetic acid, to convert hydroxyl groups into carboxymethyl groups [34]. The degree of substitution defines its resorbability [35]. In the case of OC, a functional COOH group is added to cellulose in the oxidization process by nitrogen tetroxide (N 2 O 4 ) or nitroxyl radicals, such as 2,2,6,6-tetramethyl pyperidine-1oxyl (TEMPO) [36,37]. TEMPO-oxidized cellulose has a negative surface charge introduced into the fibrils, and the consequent electrostatic repulsion forces keep the fibrils in a stable colloidal dispersion [38]. Polyelectrolyte complexes based on chitosan and cellulose represent ionic complexation, which belongs to physical crosslinking, where macromolecules can also form hydrogen bonds or hydrophobic interactions [18,39]. Regarding the incorporation of glycerol into the chitosan/cellulose materials, its utility at high concentrations between 20 and 80% is typically studied. It has been shown to improve elasticity in films [40], membranes, fibers, or food packaging, made mainly by freeze or heat drying [41,42], or solvent casting method [43]. To the best of our knowledge, there is only one study that incorporates glycerol into freeze-dried, chitosan-based, highly porous sponges. In this study, the researchers claimed that glycerol facilitated a crosslinking process via dehydrothermal treatment [44]. However, the effect of glycerol on the physicochemical properties of the scaffolds was not explored, and only dehydrothermal treatment was investigated as the initiator of crosslinking. The fabrication of a well-working and costeffective scaffold, which would initiate the host to form its own dermal and epidermal layer and accelerate healing, is still challenging in many ways. This research demonstrates the incorporation of glycerol into two types of highly porous polysaccharide complexes produced via freezedrying. The use of glycerol in these materials is rare, enabling us to provide a unique comparison of the physicochemical properties of Chit/ CMC and Chit/OC porous foams stabilized by glycerol. Notably, this stabilization is achieved without any additional crosslinking steps. Our findings revealed that glycerol significantly improved the material strength and elasticity of both complexes, influencing their physical, chemical, and biological characteristics. The addition of glycerol to the complexes resulted in a more stable structure, a reduced swelling ratio, increased hydrophobicity, and enhanced cellular processes. Furthermore, we have discovered notable differences in the effects of glycerol when added to the CMCor OC-based complex with the most significant changes observed in the Chit/CMC treated with glycerol. In the CMCbased complex compared to the OC-based complex, the presence of glycerol formed stronger hydrogen bonds, greatly enhanced elasticity, strength, stability, and reduced the release of degradation products, as well as the swelling ratio and wettability. Our research provides an alternative strategy to achieve mechanically stable and biocompatible scaffolds, eliminating the necessity of chemical crosslinking. 2. Material and methods 2.1. Materials and chemicals Chitosan from shrimp shells, 70% DDA, low viscosity (Chit, SigmaAldrich, Darmstadt, Germany), calcium‑sodium salt of oxidized cellulose – degree of oxidation 16–24% and M n =350 kg/mol (Synthesia, Pardubice, Czech Republic), carboxymethylcellulose sodium salt with M n =250 kg/mol, degree of substitution DS =0.7 (Holzbecher, Zlic, Czech Republic), acetic acid (99 %, Penta, Chrudim, Czech Republic), glycerol for molecular biology ≥99.5% (Sigma-Aldrich, Darmstadt, Germany), Dulbecco's Modified Eagle Medium (DMEM), fetal bovine K. Verˇ cim´ akov´ a et al. International Journal of Biological Macromolecules 275 (2024) 133602 3 serum (FBS), antibiotics (penicillin/streptomycin), amino acids, and Lglutamine, dimethyl sulfoxide (DMSO) (all reagents from SigmaAldrich), CellTiter-Blue® Assay (Promega, Madison, Wisconsin United States), phosphate-buffered saline (PBS, ADLAB, Poland), human lysoszyme (Sigma-Aldrich, Darmstadt, Germany). 2.2. Preparation of materials 2.2.1. Preparation of chitosan/cellulose scaffolds Chitosan (1% w/v) was diluted in a solution of acetic acid (0.1 M) and homogenized, similarly carboxymethylcellulose sodium salt CMC (0.5% w/v) or calcium‑sodium salt of oxidized cellulose OC (0.5% w/v) was diluted in ultra-pure type II water (Milli-Q®, ultrapure water-type II according to ISO 3696 being prepared on Elix 5 UV Water Purification System, Merck, Darmstadt, Germany) and homogenized with a mechanical stirrer (IKA Overhead Stirrers, Staufen, Germany). The mixture of either CMC or OC was slowly added into the solution of chitosan (Chit) in the volume ratio of 1:1 and homogenized to create material suspensions of Chit/CMC and Chit/OC. Material suspensions were placed in 48 well-plates and freeze-dried in an Epsilon 2-10D lyophilizer (Martin Christ, Osterode am Hartz, Germany) at freezing temperature − 35◦C under 1 mBar for 15 h followed by a secondary drying process at 25◦C under 0.01 mBar. For mechanical testing, each material suspension was poured into a 10 ×10 plastic plate in a volume of 50 mL and freezedried. In the case of glycerol-based scaffolds, a glycerol concentration of 0.01 g/mL was introduced into both types of material suspensions (Chit/ CMC and Chit/OC), mixed with a mechanical stirrer, and then subjected to a freeze-drying process as mentioned above. In Fig. 1, we schematically present the fabrication steps of the glycerol-containing scaffolds, and in Table 1, we show the summary of all sample compositions and their abbreviations that are used in this study. 2.3. Material characterization 2.3.1. Structure and morphology A scanning electron microscope (SEM) MIRA3 (TESCAN, Brno, Czech Republic) was used to characterize the inner morphology of scaffolds. Images were taken in a secondary electron (SE) emission mode, applying DEPTH scan mode, 10 beam density and a high voltage of 5 kV. The working distance was set to 15 mm. Prior to imaging, the samples were coated with a 4 nm layer of gold using the EM ACE 600 (Leica Microsystems, Wetzlar, Germany). The longest and the shortest dimensions of 200 pores from each sample were determined from the cross-section area using SEM images analyzed with ImageJ software. The average of dimensions from each sample was then calculated. To assess the porosity of the scaffolds, the threshold tool in ImageJ software was utilized, and five images of different cross-section areas from each sample were analyzed. 2.3.2. Dynamic mechanical analysis The tensile properties of various chitosan/cellulose scaffolds were evaluated using an RSA G2 Dynamic Mechanical Analyzer (TA Instruments Inc., New Castle, USA). 7 strips from each type were cut and gripped within geometry. The geometry defined the area of measurement: a length and width of 10 mm, with thicknesses ranging from 0.5 mm to 1.8 mm depending on the sample type. Measurements of thickness were conducted using a Digital Caliper 0–150 mm (Groningen, Netherlands). The initial mechanical test was carried out at room temperature, set at 23 ◦C. The subsequent test was performed under constant wet conditions with a PBS at 36 ◦C, within a custom-built chamber surrounding the grip. Each sample was allowed a 10-min swelling period before measurement. The stress-strain curves from the tensile tests are shown in Fig. 3. For statistical analysis, 4 repetitions of each strip type were used for the wet state and 3 repetitions for the dry state. 2.3.3. Swelling Each scaffold was weighed before immersing in a PBS solution (W d ), and then placed in glass vials containing PBS solution. The weight of the swollen scaffold (W s ) was recorded after carefully removing any residual PBS with a manual pipette (Hirschmann Labopette, Eberstadt Germany) at several intervals: 0, 1, 3, 5, 10, 20, 30, 60, 90, 120, and 180 min. For each time interval, a swelling ratio (SR), was calculated according to Eq. (1) to accurately determine the extent of swelling resulting from PBS adsorption. The samples were measured in triplicate, and the results are presented as mean ±standard deviation in the form of swelling curves. SR =Ws−Wd Wd (1) 2.3.4. Gel fraction To study the gel fraction of the scaffolds, the following experiment was used: first, the dry weights of the scaffolds were measured (W D ) and the scaffolds were then soaked in ultrapure water for 24 h. Then the scaffolds were put in dryer at 40 ◦C for another 24 h and weight again (W A ). The gel fraction was calculated according to Eq. (2), and the results are presented as mean ±standard deviation based on 3 measurements of each scaffold. GF =WA Wd ⋅100 [%](2) 2.3.5. Wettability To study the wettability of scaffolds, a 3 μ L droplet of deionized water (DI water, Spring 5UV purification system – Hydrolab, Straszyn, Poland) was applied on each surface of the scaffolds and the videos were recorded using Canon EOS 700D camera with EF-S 60 mm f/2.8 Macro USM zoom lens (EOS 700D, Canon, Tokyo, Japan). Using the gathered data, the duration it took for droplets to fully absorb into the porous scaffolds was identified as an average time from 3 measurements of 3 samples. 2.3.6. Fourier transform infrared spectroscopy To analyze the material composition of as received scaffolds, samples after gel fraction test and the character of eluates following enzymatic degradation, Attenuated Total Reflectance - Fourier Transform Infrared Spectroscopy (ATR-FTIR) was performed using Nicolet™ iS™ 5 FTIR Spectrometer (Thermo Fisher Scientific, USA) equipped with a diamond crystal. For the enzymatic degradation process, the scaffolds were immersed in vials containing 700 μ L of lysozyme solution (250 μ g/mL) to ensure complete submerging. At predetermined time points (7 and 21 days), 75 μ L of the solution was extracted and applied onto a glass slide. Fig. 1. Schematic of scaffold preparation based on chitosan and cellulose stabilized by the addition of glycerol. Table 1 The list of investigated scaffolds, their abbreviations and their material components. Scaffold composition Abbreviation Chitosan +Oxidized cellulose Chit/OC Chitosan +Carboxymethyl cellulose Chit/CMC Chitosan +Oxidized cellulose +Glycerol Chit/OC/Glyc Chitosan +Carboxymethyl cellulose +Glycerol Chit/CMC/Glyc K. Verˇ cim´ akov´ a et al. International Journal of Biological Macromolecules 275 (2024) 133602 4 This was then allowed to air dry for 1 h before undergoing FTIR measurement. Fresh lysozyme solution was then added to replenish the 75 μ L volume removed. The FTIR spectrum of the glass slide and lysozyme solution was taken and used as background. The ATR-FTIR spectra were obtained by averaging 32 scans at a spectral resolution of 4 cm −1 . These spectra, ranging from 4000 to 500 cm −1 in wavenumber, were normalized using min-max normalization, a technique available in OMNIC software (Thermo Fisher Scientific, USA). 2.3.7. Degradation measurements Human lysozyme was employed to conduct in vitro degradation studies on chitosan/cellulose scaffolds. The study started in a PBS solution at a physiological pH of 7.4 and a temperature of 37 ◦C. Following a 30-min swelling period, the excess PBS was removed from the vials. The samples were subsequently weighted and placed into an enzymatic solution containing lysozyme in a concentration of 250 μ g/mL. After each time interval: 0, 1, 3, 5, 7, 12, 16, and 21 days - the excess PBS was removed, followed by weighing the samples. Additionally, the percentage of mass loss (MS), was calculated using Eq. (3). MS =100 −(Wi⋅100 Ws)[%](3) where W s represents the weight of the scaffold after 30 min of swelling and Wi represents the weight of the degraded scaffold. Five measurements were recorded for each sample type, and the results are presented as mean ±standard deviation. A parallel degradation study, identical in procedure but without the enzyme was conducted in a cell culture medium in the following interval: 0, 1, 3, 7, 14, and 21 days. 2.3.8. In-direct cytotoxicity testing The potential cytotoxicity of chitosan/cellulose scaffold degradation products was tested by evaluating the viability of mouse fibroblast cells line NIH 3 T3 (Sigma-Aldrich, Darmstadt, Germany) cultured in eluates from scaffolds taken after 24 and 72 h. First, the scaffolds were inserted into a 24-well tissue culture polystyrene (TCPS) plate and 1 mL of complete culture medium (DMEM supplemented with 10% FBS, 2% antibiotics, 1% amino acids and 1% glutamine) was added into each well. All scaffolds were placed into incubation (37 ◦C, RH =90%, and 5% CO 2 ) and after 24 and 72 h, the eluates from the material were taken. Meanwhile, the mouse fibroblasts were seeded on TCPS at a density of 2 ×10 4 cells per mL in a complete cell culture medium and incubated. After 24 h of initial cell adhesion and spreading, the cell medium was removed and replaced with the eluates from all scaffolds corresponding to specific time points that were added into the cell culture. As a control to scaffold eluates, the complete medium was used, and for inducing cytotoxic conditions complete medium supplemented with 20% of DMSO was utilized. After 24 h, an assessment of the potential cytotoxicity of the scaffolds was conducted. The eluates from scaffolds were removed and replaced with a medium containing 20% of CellTiterBlue® reagent. Following 4 h of incubation, from each well, 100 μ L of media with reagent was transferred to a 96-well plate in triplicates and fluorescence was read at 560/590 nm using the microplate reader GloMax® Discover System (Promega, Madison, Wisconsin United States). 2.3.9. Direct in vitro cell-material interaction Prior to the in vitro experiments scaffolds were placed in the 24 wellplates sterilized for 30 min in UV light. Fibroblast cells line NIH 3 T3 were seeded on scaffolds in the density of 4 ×10 4 cells per mL and incubated for up to 14 days in controlled conditions (37 ◦C, RH =90%, and 5% CO 2 ). Cell viability was assessed after 1, 3, 7 and 14 days of incubation. Cell culture medium was exchanged every 3 days. To ensure the evaluation of cells growing only within the scaffold structure at each time point, samples were transferred to the new cell culture plate and incubated for 4 h in a medium containing 20% of the CellTiter-Blue® reagent. Afterwards, 100 μ m of media with reagent was transferred to a 96-well plate in triplicates and fluorescence was read at 560/590 nm using the microplate reader. Cells growing on TCPS were considered as a control. 2.3.10. Statistical analysis Data were evaluated by performing a one-way analysis of variance (ANOVA) using OriginPro. The level of significance was set at * p <0.05, ** p <0.01 and *** p <0.001. 3. Results and discussion 3.1. Scaffold's morphology Scaffolds require a porous architecture with high interconnectivity to enable cell infiltration, nutrient flow, and material integration within the host tissue. Freeze-drying is a well-known approach to create highly porous scaffolds, which can be observed in the morphology of the samples prepared in this study, Fig. 2A-D. Incorporating glycerol into the Chit/OC and Chit/CMC scaffolds did not significantly affect their pore sizes and porosity. The average pore sizes are distributed within a range of 152 to197 μ m, see Fig. 2E. The measured pore size value for the Chit/OC sample was 152.0 ±70.0 μ m, and with the addition of glycerol, it exhibited 165.7 ±62.1 μ m indicating no significant difference between the average values. For the Chit/CMC, sample the average pores size increased to 196.6 ±66.4 μ m, however the decrease was observed for the sample with glycerol having an average value of 179.5 ±64 μ m, with no significant difference between the values. The average porosity for samples: Chit/OC, Chit/OC/Glyc, Chit/CMC and Chit/CMC/Glyc were 67.5 ±7.0%, 71.6 ±10.3%, 69.3 ±10.0% and 64.1 ±9.0%, respectively, see Fig. 2F. Based on the presented measurements of pore sizes and overall foam porosity, we concluded that the addition of glycerol did not affect the morphology of the samples. The slight variations in reported values were not statistically significant. Previous studies have shown that glycerol can regulate pore size, acting as a cryoprotectant that controls ice crystals growth in freeze-dried scaffolds based on gelatin to obtain smaller, elongated and more regular pores [45]. We demonstrated that adding glycerol to freeze-dried foams of chitosan-cellulose complexes did not affect the morphology or porosity of the samples. 3.2. Scaffold's mechanical properties Mechanical properties are critical in engineered scaffolds to avoid failure of newly formed tissue due to scaffold deformation. Introducing glycerol to the Chit/OC and Chit/CMC scaffolds significantly changed their elasticity and strain, as observed under both dry and wet conditions. The average values from mechanical measurements are presented in Table 2 and representative stress-strain curves in Fig. 3. The incorporation of glycerol improved the elongation of the Chit/OC scaffold, increasing it over sevenfold (from 1.74 to 14.4%) in the dry state; Fig. 3A, and from 46.6 to 57.6% under wet conditions; Fig. 3B. Regarding the Chit/CMC scaffold, glycerol significantly improved elongation from 6.4 to 33.8% in the dry state, and from 36.5 to 53.3% in the wet state. In this context, chitosan also acts as a plasticizer, and its interaction with water significantly supported the elongation, especially in comparison to the dry state. Chitosan plasticization triggered by water has already been investigated [46]. Glycerol enhanced the strength of the Chit/CMC scaffolds in the dry state by almost 50% (from 306.3 to 461 kPa), while, surprisingly, the strength of Chit/OC decreased rapidly by 68% (from 137.5 to 40.2 kPa). However, the maximum strength increased again with glycerol addition for both types of chitosan/cellulose scaffolds under wet conditions, fiftyfold for Chit/ CMC (from 3 to 90.8 kPa) and only threefold for Chit/OC (from 8.1 to 17.9 kPa). This effect is favorable and more relevant, as scaffolds are hydrated within the in vivo environment. The observed enhancement in the results was anticipated, as glycerol is known to weaken the strength K. Verˇ cim´ akov´ a et al. International Journal of Biological Macromolecules 275 (2024) 133602 5 of intermolecular bonds among adjacent chains, thereby increasing their mobility. This leads to a notable improvement in scaffold flexibility [47]. This effect of glycerol on mechanical properties was confirmed in a few studies involving chitosan and cellulose films and membranes [40,48]. However, the preparation of these films and membranes is typically accompanied by a high temperature, which also highly supports cross-linking processes. Our results indicated the enhanced mechanical stability for Chit/CMC/Glyc foams with higher tensile strength and strain without any elevated temperature or chemical cross-linking as in previous reports [13,49–51]. That, of course, decreases the manufacturing costs. In summary, we show that the addition of glycerol yielded the mechanical performance of freeze-dried scaffolds without the necessity of a chemical cross-linking step. 3.3. Swelling, gel fraction and wettability The adequate water adsorption capacity of the scaffolds maintains the moist environment of the wound, which is crucial for the repair of skin tissue. Tailoring this function is necessary to enhance cell adhesion and proliferation. All chitosan/cellulose scaffolds absorbed water within the first 30 min and then maintained equilibrium from 30 to 180 min; Fig. 4A. The addition of glycerol to the Chit/CMC and Chit/OC significantly decreased the swelling ratio. In a different study related to chitosan films, the addition of glycerol increased the swelling ratio due to the higher presence of hydroxyl groups in the structure [52]. Another study has shown that glycerol contributed to swelling-resistance and anti-hydration ability in the air, because of the strong hydrogen bonding interactions between water and glycerol, which help lock the water inside the polymer network. This result was highly desirable because, after swelling, the strength and overall toughness of the hydrogels are typically significantly reduced [53]. The hydrogen bonds endowed hydrogels with enhanced mechanical properties and suppressed the swelling of hydrogels, resulting in good swelling resistance [54]. Various studies on chitosan/cellulose membranes have shown that the material's ability Fig. 2. SEM micrographs of the internal structure of the freeze-dried scaffolds made of A – Chit/OC, B – Chit/OC/Glyc, C – Chit/CMC, D – Chit/CMC/Glyc. E – The graphical representation of the mean average pore sizes in scaffolds. F – The calculated from 2D images porosity values for all the scaffolds. Table 2 Mechanical properties of chitosan-cellulose foams with average values of stress and strain from tests in dry and wet conditions with the standard deviation. Sample Measurement in dry conditions Measurement in wet conditions Stress [kPa] Strain [%] Stress [kPa] Strain [%] Chit/OC Chit/OC/Glyc Chit/CMC Chit/CMC/Glyc 137.5 ±29.8 40.2 ±12 306.3 ±40.7 461 ±88.4 1.74 ±0.47 14.4 ±1.5 6.4 ±0.3 33.8 ±3.5 8.1 ±1 17.9 ±4.1 3 ±1.3 90.8 ±19.1 46.6 ±1.8 57.6 ±4.6 36.5 ±5.2 53.3 ±2.5 Fig. 3. Representative stress-strain curves from the tensile test of all freeze-dried scaffolds, where A – measurement performed in dry conditions, B – measurement performed in wet conditions. K. Verˇ cim´ akov´ a et al. International Journal of Biological Macromolecules 275 (2024) 133602 6 to absorb moisture from the environment decreases as the glycerol content increases [55]. In the present work, a similar effect was observed with the addition of glycerol, showing the most significant differences between the Chit/CMC and the Chit/CMC/Glyc (Fig. 4B). The Chit/ CMC/Glyc floated on the water's surface in the vial, whereas the Chit/ CMC promptly submerged. After swelling, the scaffolds were removed from the vials and photographed immediately. This demonstrated that glycerol could retain water within the structure and prevent its release into the surrounding environment, and it supports anti-hydration ability in the air [53]. In Fig. 4C we show the gel fraction analysis, indicating the same trend as occurred in the swelling ratio. Glycerol reduced the gel fraction of the Chit/OC after 24 h of testing in aqueous solution from 84% ±0.26% to 58% ±0.58%, and for Chit/CMC from 84% ±0.46% to 62% ±0.28%. However, the gel fraction content may be reduced during the incubation and washing steps, as unbonded components are removed. Notably, the gel fraction enables us to verify the initial stability of the scaffolds concerning the crosslinking points. This indicates that glycerol, functioning as a plasticizer, influences the intermolecular interactions within the polymer network. Consequently, it reduces the network's density by forming new physical hydrogen-bond interactions [56–58]. Fig. 4D shows the duration of the complete water droplet adsorption into the scaffolds for all tested samples. Additionally, Fig. 4E provides a visual representation of the most notable time difference (ranging from 8 to 360 s) in water droplet adsorption between the Chit/ CMC and Chit/CMC/Glyc scaffolds. No significant difference was recorded in the adsorption of water droplets for the Chit/OC and Chit/ OC/Glyc scaffolds. Glycerol affects the interactions between the ionically charged groups of chitosan and cellulose (mainly CMC) and likely strengthens the creation of hydrogen bonds. Generally, after swelling, the pore size and porosity increase, maximizing the internal surface area of the scaffolds. In that case, there is a larger surface area/volume ratio, which allows cell infusion, as well as maximum cell growth [59]. The chitosan/cellulose scaffolds seem to exhibit a strong increase of the pores sizes and surface area after swelling, thus most likely allowing cells to pass through the 3D structure. The addition of glycerol has enhanced the hydrophobic properties of both chitosan/cellulose scaffolds, resulting in a significant reduction in swelling. This change appears to be beneficial, as it has led to improved cell proliferation; see Section 3.5. For scaffold engineering, high-swelled materials can be contra-productive due to enormous water uptake capacity following insufficient cell adhesion. For example, the CMC causes high swelling, which leads to low mechanical stability, and therefore its utility is limited. Typically, CMC serves as a temporary dressing, fostering a moist environment, absorbing wound exudates, or halting severe bleeding in wounds [60,61]. Modifying CMC with glycerol can potentially expand its range of applications and improve cellular processes. 3.4. Chemical composition and stability of scaffolds Incorporating glycerol into both the Chit/CMC and Chit/OC scaffolds has resulted in noticeable differences in the FTIR spectra. The increase in the intensity of the bands at 3328 cm −1 indicates a predominance of hydroxyl groups from glycerol, see Fig. 5A. These groups can form intermolecular hydrogen bonds with the functional COOH and NH 2 groups of the cellulose and chitosan chains. The presence of a hydrogenbonded OH band is affirmed through a broad band observed between Fig. 4. A – Swelling behavior of all scaffolds. B – Two representative vials with a visual demonstration of swelling, samples marked with circles: Chit/CMC sample submerged in the PBS (blue circle) and Chit/CMC/Glyc sample floating on the surface (green circle). The arrows lead to photo of scaffolds taken from vials directly after swelling. C – Gel fraction analysis in aqueous solution after 24 h. D – Time after which the water droplet is absorbed into the scaffold. E – Photographs of water droplet shape changes on Chit/CMC and Chit/CMC/Glyc scaffolds over time. K. Verˇ cim´ akov´ a et al. International Journal of Biological Macromolecules 275 (2024) 133602 7 3500 cm −1 and 3100 cm −1 . The characteristic bands of amide I at 1656 cm −1 were assigned to CO stretching and amide II at 1414 cm −1 and 1566 cm −1 were assigned to CO stretching of carboxylic acid (HCOO − ) bonds including CN vibration and NH bending from cellulose and chitosan [43,62,63] depicted in more detail in Fig. 5B. The presence of glycerol is marked by a lower intensity of bands at 1414 cm −1 and 1566 cm −1 . This could indicate that the quantity of functional groups associated with the molecular bond diminishes due to the plasticizing effect of glycerol as the plasticizers are known to weaken the strength of intermolecular bonds between polymer chains [28,64]. There is also a spectrum region from 1072 cm −1 to 1033 cm −1 that belongs to the CO vibration of cellulose and chitosan. In the presence of glycerol, a characteristic band at 1072 cm −1 disappeared, which can indicate the weakening of the CO bonding as the effect of plasticizers too. It appears that this band has shifted towards a higher wavenumber, with a slight elevation in the shoulder of the band at 1112 cm −1 . This change is due to the emergence of new CO stretching. A similar increase in the shoulder around this wavenumber, attributed to CO stretching, was observed in another study involving the addition of glycerol to chitosan-based hydrogels [62]. Another suggestion of the rise of the band at 1112 cm −1 could be due to the change in the length of the bond. If the bond length decreases, the band wavenumber shifts to higher values. Changes in bond length occur as the result of a change in the electronegativity of the neighboring atom. This is related to hydrogen bonding, where the shift to a higher wavenumber after glycerol was added can indicate the promotion of the hydrogen bonding interaction among chitosan, cellulose, and glycerol [65]. In addition, there are two new bands at 925 cm −1 and 854 cm −1 (Fig. 5B) in the presence of glycerol, which belong to CO stretching or OH deformation vibrations [66,67]. Furthermore, the presence of an alkyl chain is evident in the region of 2923 cm −1 and 2878 cm −1 , corresponding to the axial deformation of CH bonds in secondary (CH 2 ) and primary (CH 3 ) carbon atoms. In more detail, this region is depicted in the red area; Fig. 5C. This observation is consistent with findings reported in previous studies [43,65]. In Fig. 5D, there are FTIR spectra of scaffolds after gel fraction analysis. Based on the spectra, it seems that glycerol forms hydrogen interactions mainly with the amine (difference in intensities), which supports the claim of the other studies [43,57,58], that the glycerol changes hydrogen-bonding interactions involving the amino and amide groups of chitosan and made the chitosan chains more mobile. The disappearance of the CO bands around 1000 cm −1 and the decrease in intensity of OH bands could indicate that the part of unbounded glycerol, as well as parts of unbounded cellulose, seem to be washed out during the first 24 h in an aqueous solution. The density of the network is, therefore, reduced as non-crosslinked parts are being washed out. Based on degradation studies, swelling and wettability, the bonds within the polymer network after glycerol addition are relatively strong and hydrophobic, and so the stability of scaffolds increases. FTIR measurements were once again used to analyze the content of the eluates from the degraded scaffolds, collected after 7 and 21 days. The FTIR spectra of these eluates are displayed in Fig. 6A-B. The degradation process occurred in a solution containing lysozyme, a nonspecific enzyme prevalent in all mammalian tissues, known to effectively degrade chitosan. Meanwhile, cellulose underwent hydrolytic degradation. FTIR spectra have confirmed that glycerol acts as a material reinforcement agent and stabilizer, as it reduces the excretion of degraded products. This effect is attributed to the formation of hydrogen bonds between functional groups of chitosan and cellulose. The Chit/ OC/Glyc and the Chit/CMC/Glyc scaffolds successfully prevented the reduction of degraded products until day 7. There is a characteristic vibration of CO bonds and NH bonds between 1657 cm −1 and 1392 cm −1 in all eluates without glycerol on day 7, proving mostly the presence of oligosaccharides from degraded chitosan; Fig. 6A. Remarkably, the most significant difference showed the Chit/CMC/Glyc scaffold, Fig. 5. A – The overview of FTIR spectra of all scaffolds. B zoom into the green dashed rectangular region and C – The zoom into red dashed rectangular region showing changes in FTIR spectra after glycerol addition. D –FTIR spectra of all scaffolds after gel fraction analysis. K. Verˇ cim´ akov´ a et al. International Journal of Biological Macromolecules 275 (2024) 133602 8 where the reduction of the degraded products was completely halted until day 21; Fig. 6B. There is a characteristic vibration of CO bonds and NH bonds between 1645 cm −1 and 1030 cm −1 in all eluates from day 21, except eluate from the Chit/CMC/Glyc scaffold. Degradation products of chitosan are non-toxic oligosaccharides, which are either excreted or incorporated into glycosaminoglycans and glycoproteins afterwards in the metabolization pathway of the organisms [68]. The spectra of cellulose and chitosan are very similar because of the same functional groups except amino groups of chitosan and the carboxylic group in cellulose, which are overlapped. Characteristic peaks at 912 cm −1 and 756 cm −1 belong to the glass microscope slide. The spectrum of enzyme and the spectrum of a microscopic glass slide, used as a background is shown in the Supporting Information, see Fig. S1 A, together with FTIR of pure compounds Chit, CMC and OC, see Fig. S1 B. The FTIR spectra of the eluates also aligned with the degradation study, showing that the addition of glycerol significantly enhanced the stability of the Chit/CMC scaffold, in both enzymatic medium and cell culture medium, Fig. 7. Both degradation studies and the collection of eluates were conducted over a period of up to 21 days. This duration was sufficient to observe significant variations in the behavior of samples incubated in enzymatic solution and cell culture medium [69]. The most notable difference in the degradation profile was observed between the Chit/CMC and the Chit/CMC/Glyc scaffolds. In the enzymatic medium, after 21 days, Chit/CMC experienced a degradation of 94%, whereas Chit/CMC/Glyc exhibited only a mass loss of 40%. Chit/OC lost 73% of its mass, in contrast to Chit/OC/Glyc, which lost 60% of its mass, as shown in Fig. 7A. In the cell culture medium, Chit/CMC lost 59% of its mass, while Chit/CMC/Glyc showed a significantly lower mass loss of just 22% at the end of the 21-day period. During the entire degradation time, the difference in mass loss between Chit/OC and Chit/OC/Glyc was negligible, with both scaffolds reaching 45% and 48% of mass loss, respectively, see Fig. 7B. Our findings align with those of a similar study that explored the complexation of non-crosslinked, freeze-dried chitosan and CMC. In that study, enhanced long-term stability was achieved through thermal treatment. Their degradation study in a biofluid, akin to our degradation study in a cell culture medium, revealed that the most stable scaffold incurred only 20% mass loss at 28 days. However, it is important to note that the concentration of chitosan used in their study was significantly higher than in our scaffolds [70]. Additionally, when comparing the degradation results of our Chit/CMC/Glyc with other studies, where enzymatic degradation [69,71–73] or hydrolytic and medium degradation [13] were provided on crosslinked chitosanbased samples, or cellulose-based scaffolds degraded in a hydrolytic environment [51], the stability of Chit/CMC/Glyc appears to surpass or reach similar levels of mass loss. Importantly, this observation suggests that the use of a chemical crosslinking agent between chitosan and cellulose might be unnecessary because similar prolonged stability of the scaffolds can be achieved by glycerol addition prior to freeze-drying. The mechanism of interactions between chitosan, cellulose in both variations (OC and CMC) and glycerol is schematically presented in Fig. 8. The variations in the impact of glycerol addition into the Chit/CMC and Chit/OC complexes may stem from the differences in their salt content; CMC contains monovalent salts (Na), whereas OC additionally includes multivalent salts (Na and Ca). Multivalent cations tend to interact more strongly within the chain compared to monovalent cations, leading to intramolecular clustering within charged polymer chains or potentially causing dimerization between acid units or the formation of mixed acid-salt dimers [74]. This chain conformation can result in the shrinkage of polymer chains, which could hinder the formation of hydrogen bonds as indicated in the schematics in Fig. 8. Previous studies have confirmed that multivalent cations strongly influence chain conformation, with chain shrinkage being more pronounced in the presence of multivalent ions than in the presence of monovalent ions [75,76]. The bonding of cations occurs through ionic interaction, which is generally stronger than hydrogen bonds; thus, our hypothesis is that the ionic bonds potentially override the hydrogen bonding of glycerol. This elimination of spaces for hydrogen bonding might account for the notably lower stability of the Chit/OC/Glyc scaffold and its accelerated release of degradation products in comparison to the Chit/CMC/Glyc scaffold. 3.5. Biocompatibility studies At first, we performed an indirect experiment to verify if the degradation products of scaffolds cause any in vitro cytotoxic response of cells. As we showed in the degradation study (Fig. 7B), there is a notable mass loss for all scaffolds after 1 day of incubation and increased after 3 days. Nevertheless, the cytotoxicity test showed that the cellular viability is not reduced by >20% compared to the control, ranging from 82% to 98%, which is within the range of biocompatible materials; Fig. 9A. We observe lower cell viability when exposed to eluates after 3 days of scaffolds incubation in cell culture medium, which can be attributed to the higher content of degradation products. Experiments that involved direct cell seeding on both Chit/OC/Glyc and Chit/CMC/ Glyc scaffolds have shown a significant increase in cell proliferation; Fig. 9B. We claim that glycerol changed the physical-chemical properties of material, which facilitates the cell processes in the 3D structure of the scaffolds. As the maximum strength and the elasticity improved with glycerol addition, the stability was also prolonged, as well as hydrophobic nature was induced into chitosan/cellulose, which triggers cellular processes. However, the results from direct cell seeding are modest compared to the control, which is the surface of the 2D plastic well plate, where cells adhere and grow more rapidly than in 3D porous materials. To highlight the glycerol effect, the yellow arrow in the graph in Fig. 9C indicates the same experimental results without control for clarity. Comparing cell interactions with glycerol is challenging, as most of the literature focuses on glycerol as a cryopreservant for cells and tissues [77]. Glycerol is known to reduce water crystallization, inhibit ice crystal growth, and, therefore, reduce damage to cell structure and Fig. 6. The FTIR spectra of eluates from degraded scaffolds collected on a microscopic glass slide; A – collected after 7 days, B – collected after 21 days. K. Verˇ cim´ akov´ a et al. International Journal of Biological Macromolecules 275 (2024) 133602 9 function [78]. However, studies have reported glycerol-induced cytotoxicity at higher concentrations (10% v/v) in HaCaT keratinocytes [79], primary murine cell cultures [80], and inhibitory effects on the proliferation of the Baby Hamster Kidney (BHK) cell line [81]. Conversely, other studies suggest beneficial effects, such as glycerol aiding in maintaining fibroblast proliferation under oxidative stress [82], or the use of cellulose glycerol-based membrane as non-cytotoxic, supporting long-term cell adhesion, spreading, and proliferation in both NIH3T3 (Mouse embryonic fibroblast cell line) and HaCaT (human keratinocyte cell line) [83]. In summary, the impact of glycerol, particularly in materials, on cellular processes still requires further research. 4. Conclusions The addition of glycerol can significantly enhance the physical and biological functions of polysaccharide complexes based on chitosan and cellulose foams. The properties improvement was achieved without compromising the high porosity of the freeze-dried scaffolds, which was above 64% for all studied samples. Moreover, it improves stability by slowing down the degradation. After 21 days of incubation in enzymatic medium samples containing glycerol exhibited mass loss below 60%, while the ones without plasticizer degraded over 76%. This, in turn, reduces the release of degradation products. Moreover, mechanical properties can be tailored by the addition of glycerol. We observed a 50% increase in strength and over fivefold higher elongation for the Chit/CMC/Glyc scaffold compared to the sample without glycerol. Importantly, we proved that additional crosslinking agents are unnecessary, and the fabrication process can be simplified to just one freezedrying step, reducing the manufacturing costs. The interactions between glycerol and the Chit/CMC and Chit/OC scaffolds are based on physical interactions between polysaccharide charge complexes. Importantly, the interactions between CMCand OC-based scaffolds differ due to the distinct presence of multivalent cations in CMC and OC. In CMC-based scaffolds, glycerol forms stronger and more easily facilitated hydrogen bonds with the CMC chain. In contrast, OC-based scaffolds exhibit a preference for ionic interactions, leading to chain shrinkage and reduced accessibility for hydrogen bonding. Biologically, glycerol was found to be non-cytotoxic and significantly contributed to an increased fibroblast proliferation rate, which was four times higher in the Chit/OC/Glyc sample compared to the scaffold without glycerol after 21 days of incubation. This effect is correlated to improved mechanical properties, enhanced stability, and increased hydrophobicity, which in turn reduces swelling of scaffolds. Glycerol appears to strike an optimal balance between hydrophilic and hydrophobic environments, thereby promoting cell attachment and proliferation. We clearly showed that glycerol has the ability to enhance both the biological and physical functions of chitosan and cellulose, improving their biocompatibility for skin tissue engineering. The freeze-dried foams based on chitosan and cellulose are believed to support the mechanical and biological protection of the wound bed and avoid possible cytotoxicity of crosslinkers. The addition of glycerol also minimizes the production time and reduces the production costs. Further, the scaffolds have a great potential to be enriched with additional proteins or other active compounds in one preparation step for different medical applications. Supplementary data to this article can be found online at https://doi. org/10.1016/j.ijbiomac.2024.133602. Ethics and availability Ethics approval was not required for this research. CRediT authorship contribution statement Katarína Verˇ cim´ akov´ a: Writing – review & editing, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Joanna Karbowniczek: Writing – review & editing, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Marian Sedl´ aˇ r: Investigation. Urszula Stachewicz: Writing – review & editing, Resources, Funding acquisition. Lucy Vojtov´ a: Writing – review & editing, Supervision, Resources, Fig. 7. The degradation of all scaffolds over 21 days of experiments in A – enzymatic medium. B – cell culture medium. Fig. 8. The schematics of possible polysaccharide charge complexes in the presence of multivalent ions in Chit/OC/Glyc compared to monovalent ions in Chit/CMC/Glyc, resulting in a chain shrinkage due to dimerization and elimination of spaces for hydrogen bonding. K. Verˇ cim´ akov´ a et al.