Comprehensive Characterization of Solution‐Cast Polycaprolactone/MXene/Gelatin Composite Films for Biomedical Applications

ABSTRACT

Despite significant advances in the development of biocompatible platforms, such as scaffolds, films, and hydrogels, a challenge remains in formulating films with the right balance of mechanical properties and bioactivity. Herein, we developed biocompatible composite films based on polycaprolactone (PCL), MXene, and gelatin that can be utilized for biomedical applications. PCL and gelatin (from bovine, fish, and porcine skin) were used to design the biocompatible matrix, while MXenes were used as a filler to enhance the mechanical and biological properties of the films. We investigated the influence of these three types of gelatin on the chemical structure, morphology, physicochemical properties, cytotoxicity, biocompatibility, and cell growth. All the films exhibited high tensile strength, ranging from 5 to 10 MPa. The incorporation of a relatively small content of MXene (0.5 wt%) altered the tensile properties of the films with the lower gelatin contents (12–15 wt%). SAXS analysis revealed that the nanometer‐scale lamellar stack structures characteristic of PCL, consisting of alternating crystalline and amorphous lamellae, were present in all samples and exhibited a morphology identical to that of neat PCL. In contrast, WAXS showed that the relative intensities of individual PCL reflections varied with sample composition, indicating a preferential orientation of PCL crystallites—and consequently of the lamellar stacks—particularly, in MXene‐containing samples. The SEM/SE micrographs displayed a coarse morphology of gelatin nanoparticles in the PCL matrix, and the structure coarseness decreased in the following order: PCL/MX/fish gelatin > PCL/MX/bovine gelatin > PCL/MX/porcine gelatin. In vitro, cell culture experiments with SAOS‐2 cells revealed that cell confluence was relatively high on samples with the 14.5 wt% porcine gelatin.

Polycaprolactone (PCL)/Gelatin/MXene composite films design and evaluation, where blending PCL and natural gelatin (bovine, porcine, and fish) and MXene nanoparticles resulted in coarse morphology, enhanced mechanical strength, and tunable cell‐material interactions, highlighting porcine‐based composite as a promising material for biomedical applications.

1. Introduction

Developing biomaterial platforms such as hydrogels, films, or meshes for tissue engineering is challenging, as the biocompatibility and bioactivity of the material can influence many functions of the cells, including anchorage [1], signaling [2], and morphogenesis [3]. Moreover, these biomaterial platforms provide physical support for tissues [4], which is crucial for all these processes. Recently, biodegradable polymers such as chitosan [5, 6, 7], alginate [8], cellulose [9, 10], hyaluronic acid [7, 11], gelatin [12, 13], collagen [5], and silk [8], have been extensively used to design the biomaterial platforms. Among them, gelatin is increasingly used in biomedical applications due to its hydrophilicity, biocompatibility, biodegradability, film‐forming ability, low production cost, and structural similarity to native tissues [14, 15, 16]. Gelatin is a substance derived from collagen through hydrolysis and may originate from various animal sources, including bovine [17, 18], porcine [17], and fish [19, 20]. Collagen hydrolysis is carried out in either acidic or alkaline media, resulting in type A and type B gelatin, respectively [21]. Acidic treatment is typically employed for collagen with a lower degree of covalent crosslinking, such as collagen derived from fish or porcine skin, whereas alkaline treatment is used for highly crosslinked collagen from bovine skin [22].

The amino acid composition and sequence of gelatin vary depending on its source; however, they are characterized by high contents of glycine, proline, and hydroxyproline [23]. Various methods have been developed to distinguish gelatin obtained from different sources, such as enzyme‐linked immune sorbent assay (ELISA) [24], gel electrophoresis [25], polymerase chain reaction [26], liquid chromatography [27], and mass spectrometry [28]. In a recent study, Hassan et al. utilized a hydrophilic interaction liquid chromatography coupled with tandem mass spectrometry (HILIC/MS) method and partial least squares‐discriminant analysis (PLS‐DA) to effectively distinguish porcine, bovine and fish gelatin sources [29]. Moderate differences in amino acid composition were observed among the three gelatin types [30]. Notably, cold‐water fish gelatin, which was used in this study, exhibits greater variation, particularly in the content of proline and hydroxyproline (15%–17%). In contrast, bovine, porcine, and warm‐water fish gelatins contain higher combined levels of proline and hydroxyproline (20–22 mol%). These amino acids play a critical role in stabilizing the triple‐helix structure, thereby contributing to increased gel strength and enhanced mechanical robustness.

The collagen‐like triple‐helix structure is the primary architect of the functional properties of gelatin. Its presence is directly correlated with the gel strength (bloom strength). In other words, the greater the extent of renaturation toward a collagen‐like structure, the mechanically stronger the gel becomes [31]. The triple‐helix structures act as physical junction zones, which melt upon heating just below human body temperature [32], giving gelatin its unique melt‐in‐the‐mouth characteristic. The physicochemical properties of gelatin are strongly affected by many factors, such as amino acid composition, isoelectric points, hydrophobicity, presence of low molecular weight protein fragments, and the difference in the molecular weight distribution, which result from variations in the extraction conditions [33, 34]. Recent works on origin‐dependent molecular ordering in gelatin or gelatin characteristics further demonstrate that gelatin from different animal sources (bovine, porcine, fish, etc.) and/or different body parts (skin, bones, etc.) exhibited distinct patterns of chain packing, hydrogen bonding, supramolecular organization, and characteristics even when their overall amino acid compositions are comparable. Additionally, these molecular and structural features can be significantly influenced by the method of gelatin preparation from collagen [35]. The origin‐dependent differences in molecular ordering lead to source‐specific variations in viscosity, fiber morphology, and mechanical properties [22, 36, 37].

Gelatins from various sources may differ in the cell response to them [38, 39, 40]. In composites, gelatin can also influence the mechanical properties of the resulting material due to differences in its molecular structure or through different emulsification in the matrix [41]. Previous studies have reported the influence of gelatin sources on the rheological properties of GelMA for bone regeneration [42]. However, due to its low mechanical strength and rapid degradation rate of gelatin, it is usually blended with other polymers or incorporated with inorganic nanoparticles to improve its mechanical properties [43, 44]. To overcome this challenge, various strategies have been adopted, for example, grafting, blending with other polymers or incorporating nanoparticles.

In a recent study, DNA was grafted onto gelatin to enhance the pharmacodynamics efficacy of antibiotics [45]. Sun et al. developed nanocomposites by incorporating silver (Ag) nanoparticles into gelatin, which have shown potential to act as anticancer agents in breast cancer [46]. Hemdan et al. modified the strategy and functionalized gelatin films with silver‐doped zinc oxide nanoparticles to develop a multifunctional wound dressing for infection control [14]. Darban et al. incorporated gelatin hydrogel with MXenes and zingerone to develop composite hydrogels (GPM Z) for wound dressing [47]. In vitro studies indicated that antibacterial evaluation against Escherichia coli and Staphylococcus aureus demonstrated that the antibacterial property of MXenes' sharp‐edged nanosheet structure and antioxidant properties of zingerone significantly enhanced wound healing. An in vivo study in a rat model demonstrated that GPM‐5Z effectively prevented wound contraction infections by 98.75% on Day 14. Although gelatin has excellent bioactivity, it produces films with insufficient mechanical properties which limits their biomedical applications. It has to be combined with other polymers to enhance biological and physicochemical properties [48].

Among various synthetic biodegradable polymers, PCL has been widely studied due to its slow degradation rate and mechanical robustness. It is an FDA‐approved polymer [49] which is known to enhance the osteogenic potency of human mesenchymal stem cells [50]. PCL‐based composites have been studied for wound dressing [51, 52], tissue regeneration [53, 54], and bone regeneration [55, 56]. In a recent study, PCL was incorporated with pectin‐modified halloysite for wound healing application. PCL membranes loaded with polymyxin B have been used to develop as controlled delivery systems [57]. However, its hydrophobic nature limits cell attachment and protein absorption. Gelatin, on the other hand, mediates cell adhesion, cell proliferation, and tissue regeneration [58]. Therefore, gelatin and PCL can be combined to produce composite films that offer a balance of biological and mechanical properties, making them suitable for biomedical applications. Furthermore, the incorporation of MXene nanoparticles into PCL/Gel films will further enhance their properties. The high surface area and hydrophilic nature of MXene have the ability to interact with eradicate bacteria and enhance the wound healing process [59].

Numerous studies have shown that integration of MXene nanoparticles into various matrices has enhanced tissue reconstruction due to the large surface area and adjustable physicochemical properties of these nanoparticles [60, 61, 62, 63]. MXenes are 2D materials consisting of carbides and nitrides with distribution of functional groups having the formula M _n + 1X _n T _x, where M refers to the transition metal, X—carbon/nitrogen, T—functional groups such as oxygen, hydroxyl, fluorine, and n is an integer [64]. MXenes have a high specific surface area and possess hydrophilic functional groups (such as OH, O, and F), which creates an optimum atmosphere for the cells to grow, proliferate, and differentiate [65, 66]. Other properties, such as osteogenic differentiation ability, also promote cell growth and bone regeneration [67]. Their electrical conductivity can support cell communication and differentiation, particularly in electroactive tissues, such as neural or muscular tissues [68]. A recent study showed that 1 wt% loading of MXene enhanced the tensile strength of PLA films from ~6 to 17.64 MPa and Young's modulus from ~398 to 1056 MPa [69]. Chitosan films also showed an increase in tensile strength and Young's modulus by about 303% and 830%, respectively, by a small addition (0.04 wt%) of MXenes [70]. In another study, Hui et al. noticed that 3 wt% MXene addition to PCL/CS composite increased Young's modulus from 0.4 to 12 MPa [61]. These studies indicate that MXene is a good option to enhance mechanical properties of PCL/gelatin films.

Although extensive research has explored different combinations of PCL, gelatin, and MXenes for various biomedical applications [71, 72, 73], most reports indicate the use of either PCL‐gelatin or PCL‐MXene as a matrix to design the films, nanofibres, or composites. For example, PCL/Gelatin nanofiber membranes were incorporated with liposomes of epigallocatechin‐3‐gallate (EGCG) for their use in skin regeneration [74], PCL/gelatin nanofibrous scaffolds were reinforced with layered double hydroxide (LDH) nanoclay to promote the growth of nerve cells [75], gelatin hydrogel was modified with PCL nanofibers and carbon nanofiber (CNF) to produce scaffolds for bone regeneration [76], PCL/gelatin matrices for pneumothorax treatment [77], PCL/gelatin/hydroxyapatite (HAp) composites for treatment of bone damage [78, 79], and PCL/gelatin/polyaniline scaffolds for cardiac tissue regeneration [80]. Similarly, PCL‐MXene‐based scaffolds are used in tissue engineering [72], regenerative medicine [81], repair of maxillofacial bone defects [82], and anticancer drug delivery system [83]. The unique combination of PCL, gelatin, and MXenes has rarely been utilized to form composite films. The developed materials could be utilized for wound healing or developing scaffolds or membranes for medical applications.

The combination of PCL, gelatin, and MXenes in a composite material may offer synergistic enhancement of biological performance, mechanical stability, and functional properties, which could make the materials highly suitable for biomedical applications, including tissue engineering and regenerative medicine. PCL will provide mechanical strength and biodegradability, and thus it can supply the structural support necessary for longer‐term implantation. At the same time, it can be easily processed into various shapes, for example, by electrospinning or 3D printing [84]. Gelatin will act as a biologically active component promoting cell adhesion, proliferation, and differentiation. Its hydrophilicity will also contribute to enhancement of the cell‐material interactions. The poorer mechanical properties and chemical stability of gelatin will be compensated for by PCL's presence in the composites [71, 85, 86]. MXenes will also contribute to mechanical reinforcement of the PCL‐based composite [72, 87, 88]. However, their main contribution will presumably consist of adding functional properties related to their electrical conductivity and their large specific surface area with a high density of active functional groups. The synergistic effect of the three components will thus lie in PCL providing mechanical strength and long‐term stability, gelatin offering biological functionality, and MXenes adding new functional properties. This integrated approach may thus create a material with optimized properties for tissue engineering and regeneration.

In the current study, we synthesized biocompatible films based on PCL, MXene, and gelatin (obtained from porcine, bovine, and fish) using solvent casting. The novelty of this work lies in two aspects: (a) development of composite films based on a unique combination of PCL–gelatin–MXene, which has not been reported so far. (b) Utilizing three different types of gelatin (bovine, porcine, and fish) for the preparation of these composites and comparing the properties of these films by various physicochemical techniques. We hypothesize that gelatin obtained from different sources will change the physical, chemical, and biological properties of resultant composite films, which can impact their performance for various biomedical applications. Although few studies have blended polycaprolactone/gelatin using electrospinning [89, 90], solvent casting [89], and electrospraying technology [50], they have used a single gelatin component without blending any additional nanoparticles. A comparative study of composites obtained from different gelatin sources has not been conducted. Further, we investigated the chemical structure with FT‐IR, morphology on multiple scales using small‐ and wide‐angle x‐ray scattering (SWAXS), light microscopy (LM), scanning and transmission electron microscopies (SEM, TEM), and tensile mechanical properties using static testing. SEM/energy‐dispersive analysis of x‐ray (EDX) was used to determine the distribution of MXene particles in the polymer matrix of the prepared films. Cell culture experiments were performed to test the biocompatibility of the developed films. A comparative analysis of the composite films concerning their morphology, mechanical properties, and cell growth was carried out.

2. Experimental

2.1. Materials

PCL (MW = 50,000) was purchased from Polysciences Inc. (Germany). Gelatin (Gel) from porcine skin (gel strength—300, Type A), bovine gelatin (gel strength ~225, Type B), and fish gelatin (40%–50% in H₂O) were purchased from Sigma Aldrich Co Ltd., USA. MXene (MX; formula: Ti₂C) was obtained from Laizhou Kai Kai Ceramic Materials Co. Ltd., China. Chloroform was obtained from Sigma‐Aldrich Co. Ltd., USA. All the chemicals were analytical grade and used as received.

2.2. Preparation of PCL/Gel and PCL/MX/Gel Films

PCL/Gel films (PCL/B1‐B3, PCL/F1‐F3 and PCL/P1‐P3) were prepared using one‐pot synthesis by dissolving gelatin (1.2 g) in 20 mL of acetic acid and formic acid mixture (6.1) and stirring at 60°C for 20 min. PCL (9 g) was dissolved separately in 20 mL of chloroform and stirred at 40°C for 10 min until a homogenous solution was formed. However, MXene was not easily dispersible in chloroform. Subsequently, PCL/Gel films were fabricated by mixing both solutions in different ratios, stirring for 10 min, and solution casting using a TQC automatic casting machine (TQC, Germany). The obtained films were dried at room temperature for 24 h, and the final thickness was around 300 μm.

PCL/MX/gel films (PCL/MX/B1‐3, PCL/MX/F1‐3, and PCL/MX/P1‐3) were prepared using the abovementioned procedure. Briefly, gelatin (1.2 g) was dissolved in 20 mL of acetic acid‐formic acid mixture (6.1) and stirred for 20 min. PCL (9 g) was dissolved separately in chloroform (20 mL), and MXene (0.06 g) was added to the PCL solution and stirred for 10 min until a homogeneous solution was obtained. The two mixtures of PCL/MXene and gelatin solutions were then mixed and stirred for 10 min. The films were fabricated by solution casting and dried like the PCL/gel films. The compositions of the composite films and their respective sample codes are presented in Table 1.

TABLE 1.

Composition of PCL/gel and PCL/MX/gel films.

Sample code	Dry composition (wt%)			Sample code	Dry composition (wt%)			Sample code	Dry composition (wt%)
	PCL	B_Gel	MX		PCL	F_Gel	MX		PCL	P_Gel	MX
PCL	100			PCL	100			PCL	100
PCL/B1	88	12		PCL/F1	88	12		PCL/P1	88	12
PCL/B2	85	15		PCL/F2	85	15		PCL/P2	85	15
PCL/B3	75	25		PCL/F3	75	25		PCL/P3	75	25
PCL/MX/B1	88	11.5	0.5	PCL/MX/F1	88	11.5	0.5	PCL/MX/P1	88	11.5	0.5
PCL/MX/B2	85	14.5	0.5	PCL/MX/F2	85	14.5	0.5	PCL/MX/P2	85	14.5	0.5
PCL/MX/B3	75	24.5	0.5	PCL/MX/F3	75	24.5	0.5	PCL/MX/P3	75	24.5	0.5

Abbreviations: B_Gel, bovine gelatin; F_Gel, fish gelatin; P_Gel, porcine gelatin.

2.3. Physicochemical and Mechanical Characterization

Infrared spectroscopy (FT‐IR) was used to analyze the chemical structure of the composite films. Infrared absorbance spectra were performed on a Nicolet 380 FT‐IR spectrometer ranging from 650 to 4000 cm⁻¹ using the ATR (attenuated total reflection) accessory with a Ge crystal.

The SWAXS experiments were carried out on a SAXSess mc² instrument (Anton Paar, Austria), which is a Kratky‐type instrument equipped with a microfocus x‐ray source with a Cu anode, single‐bounce focusing Xx‐ray optics, and a collimation block. Rectangular imaging plates were used for x‐ray detection. The range of the scattering vector magnitude, q, was 0.2–28 nm⁻¹; here q = 4π/λ sin (θ), λ is the wavelength, and θ is ½ of the scattering angle. The x‐ray wavelength, λ, was 0.1542 nm (CuKα line). The partly 2D (i.e., a rectangular sector) scattering patterns of the film samples were measured in the standard, that is, transmission mode in vacuum, perpendicular to the film surface. The 1D radial intensity profiles were calculated from the measured 2D patterns by azimuthal averaging using the software supplied with the instrument. This procedure could be performed due to the azimuthal symmetry of the scattering/diffraction patterns (thanks to the isotropic nature of the samples in the plane of the films). Then, the SWAXS profiles were separated into SAXS and WAXS profiles. The Irena software toolbox [1] for small‐angle scattering analysis was used to process the SAXS and WAXS profiles.

The SEM with SEM/EDX of top, bottom, and fracture surfaces was performed with a high‐resolution SEM microscope (MAIA3; Tescan, Czech Republic) equipped with an EDX detector (X‐Max^N 20; Oxford Instruments, UK). Each sample was observed at the top surface (mat and rougher), bottom surface (shiny and smoother), and at the fracture surface (fracturing in liquid nitrogen, as described elsewhere) [91]. The fracture surfaces were perpendicular to the top surface. Before observation, the samples were fixed with double adhesive carbon tape to aluminum sample holders and covered with a thin platinum layer (∼4 nm) using a vacuum sputter coater (SCD 050, Leica; Austria) to minimize possible e‐beam‐induced charging and sample damage. The surfaces were visualized with a secondary electron detector (SEM/SE) at the accelerating voltage of 30 kV, a backscattered electron detector (SEM/BSE) at 30 kV, and the elemental microanalysis from selected typical regions (SEM/EDX) was performed at 30 kV.

The TEM was done using a TEM microscope (Tecnai G2 Spirit; FEI, Czech Republic). The ultrathin sections for TEM observations were prepared by ultramicrotomy at cryogenic conditions (ultramicrotome: Ultracut UCT; Leica, Austria; sample and knife temperature during cutting: −80°C and −50°C; nominal thickness of the sections: 60 nm). The ultrathin sections were collected onto the standard copper grid, transferred to the TEM microscope, and observed using the standard bright field imaging at an accelerating voltage of 120 kV.

Tensile mechanical properties were measured using a DMA Q800 dynamic mechanical analyzer (TA Instruments, USA) at RT. The specimens had a width of 6 mm and a thickness of about 300 μm. The gauge length between the clamp jaws was 15 mm.

2.4. Cell Culture Experiments

The SAOS‐2 human osteosarcoma cell line was used for in vitro testing of the PCL‐based nanocomposite films. The films were cut with a cleaned 15 mm diameter punch by circular motion on a sterile plastic surface. The films were disinfected with 70% ethanol for 15 min, washed with phosphate‐buffered saline (PBS), and inserted in a 24‐well plate (Techno Plastic Products, Switzerland). The polystyrene (PS) surface of the tissue culture plate was used as a standard control tissue (TCPS). The cells were seeded at the density of 10,000 cells/cm² in McCoy's 5A medium (Sigma Aldrich; MA, USA) supplemented with 15% fetal bovine serum (FBS; Gibco—Thermo Fisher Scientific; MA, USA), and cultivated for 1, 3, and 7 days under standard culture conditions (37°C and a humidified air atmosphere with 5% of CO₂).

For the microscopy observation, the cells were fixed in a 4% solution of paraformaldehyde in PBS and fluorescently stained for nuclei and actin filaments using 4′,6‐diamidino‐2‐phenylindole (DAPI) and phalloidin‐tetramethylrhodamine conjugate (Sigma‐Aldrich), respectively. The cells were photographed on an inverted IX71 epifluorescence microscope equipped with a DP80 camera (both Olympus, Japan) using a 10× lens (N.A. = 0.3). Three independent experiments were performed to determine the cell number.

The metabolic activity of cells was measured by the MTS test (Promega; WI, USA) using the protocol provided by the manufacturer. Dulbecco's modified Eagle medium and 10% FBS cultivation medium were mixed with MTS reagent in a ratio of 6:1 volume/volume to obtain the working solution. The cells were cultivated in the working solution for 2 h. The solution was then transferred in triplicate to a 96‐well plate (100 μL per well), and the absorbance was measured at 490 and 700 nm as background using a VersaMax plate reader (Molecular Devices LLC; CA, USA). Cell mitochondrial metabolic activity was evaluated in three independent biological experiments (n = 3). In each experiment, each sample type was tested in triplicate wells (n = 3 per group), and absorbance was measured three times per well (technical triplicates), resulting in nine technical measurements per sample type in each experiment.

Cell nuclei were quantified from DAPI‐stained fluorescence micrographs using the Cellpose segmentation algorithm (Cellpose v4.0.8, Stringer et al. [92], see Figure S1). Images were exported as 8‐bit TIFF files without contrast enhancement prior to analysis. Due to the heterogeneous background of the substrates, segmentation was performed using the retrained “nuclei” model with manual adjustment of the diameter parameter (30 px) to match the average nuclear size of SAOS‐2 cells. No additional background subtraction was applied to avoid artificial alteration of nuclear boundaries. Segmentation masks were visually inspected for each image to exclude evident false positives originating from substrate autofluorescence or imaging artifacts. For each material type and time point, images from multiple randomly selected fields (n = 10 from 3 independent experiments) were analyzed. The total number of nuclei per field was recorded and normalized to the imaged area to obtain nuclei density (cells/cm²).

The statistical significance for both metabolic activity measurement and cell nuclei quantification was calculated by one‐way ANOVA with Holm‐Šídák post hoc test (α level = 5%) using SigmaStat 4 (Grafiti LLC, Palo Alto, CA, USA) software. The error bars in the Figures 10 and 11 indicate standard deviations.

FIGURE 10.

Metabolic activity of the SAOS‐2 cells cultivated on the studied composites and control pure PCL (PCL) and tissue culture polystyrene (PS) for 1, 3, and 7 days. *Statistically significant difference to PS in respective time interval on α level = 5% calculated by one‐way ANOVA with Holm Šídák post hoc test. The error bars indicate standard deviation.

FIGURE 11.

Cell number of the SAOS‐2 cells cultivated on the studied composites and control pure PCL (PCL) and tissue culture polystyrene (TCPS) for 1, 3, and 7 days. *Statistically significant difference to TCPS in respective time interval on α level = 5% calculated by one‐way ANOVA with Holm–Šídák post hoc test. The error bars indicate standard deviation.

3. Results and Discussion

3.1. Molecular Structure From FT‐IR

FT‐IR spectra of the composite samples exhibited contributions from both PCL and gelatin (Figure 1a,b). The characteristic peaks of gelatin were observed at 1652 and 1544 cm⁻¹ corresponding to the amide‐I [93] and amide‐II [94] bands, respectively. The amide‐I band is mainly due to C═O stretching vibrations, while the amide II band arises due to N─H stretching vibrations strongly coupled to the C─N stretching vibration of gelatin [95, 96]. The peaks observed at ~2945 and ~2866 cm⁻¹ were attributed to asymmetric and symmetric stretching of methylene groups in PCL, respectively. Further peaks assigned to PCL were observed at 1724 cm⁻¹ (C═O stretching of the ester carbonyl group), 1472 and 1366 cm⁻¹ (CH₂ bending), 1294 cm⁻¹ (backbone C–O and C–C stretching in the crystalline phase), 1242 cm⁻¹ (asymmetric COC stretching), 1188 cm⁻¹ (C–O stretching), 1173 cm⁻¹ (symmetric COC stretching), 1108; 1047; 961; and 732 cm⁻¹ (CH₂ rocking) [97, 98, 99, 100]. The FT‐IR spectra of all three composites prepared using gelatin obtained from fish, porcine, and bovine skin showed a similar spectral pattern with no significant difference. The amide I and II peaks at 1652 and 1544 cm⁻¹ are the leading indicators showing an apparent variation. These two peaks increase gradually with the increase in the gelatin concentration, indicating the presence of gelatin in the samples.

FIGURE 1.

(a) Chemical structure of PCL and gelatin; (b) FT‐IR spectra of PCL, gelatin, and PCL/F1‐F3 composites, and (c) gelation obtained from bovine, fish, and porcine.

The spectra of the bovine‐derived and porcine‐derived gelatin films were similar to those of fish‐derived gelatin samples. MXene was not detected in the FT‐IR spectra due to its low concentration with respect to the FT‐IR sensitivity.

FT‐IR spectra of the three investigated types of gelatin (Figure 1c) were very similar; there are only subtle differences among them. The assignment of the principal bands is given in Table 2. The positions of the amide I (around 1640 cm⁻¹) and amide III (around 1240 cm⁻¹) bands indicate a predominantly random‐coil structure of gelatin, possibly with some fraction of β‐sheets [106]. The occurrence of triple‐helix structure in the investigated samples was minimal or none based on the obtained FT‐IR spectra. Even though the amino acid compositions of bovine, fish, and porcine gelatins differ, it is obvious from the shown spectra that a simple visual comparison of FT‐IR spectra of the investigated samples was not enough to differentiate between them and to reliably determine possible structural differences. More involved techniques have to be applied to the amide I, amide II and amide III band regions to discriminate between these three different types of gelatin, such as hierarchical cluster and principal component analysis using [107], advanced pattern recognition [108] or fuzzy graph‐based chemometrics [109].

TABLE 2.

Assignment of the principle bands of gelatin.

Band maximum (cm−1)	Assignment	References
3300	Amide A band (NH stretching)	[101]
3074	Amide B band	[101]
2940	CH2 stretching	[102]
1640	Amide I band	[101, 103]
1540	Amide II band	[101, 103]
1450	CH2 bending (scissoring mode)	[103, 104]
1336	CH2 bending (wagging mode)	[103, 104]
1240	Amide III	[101, 103]
1082	C–O stretching in amino acid side chains	[103, 105]

3.2. Nanoscale Morphology and Crystal Structure From SWAXS and XRD

PCL is a semicrystalline polymer containing lamellar‐stack structures with alternating crystalline and amorphous lamellae [2, 3]. All the reflections in the WAXS profiles of the selected samples (Figure 2) were assigned to the orthorhombic crystal lattice of PCL [4]. The profile of the initial PCL powder sample is shown for comparison.

FIGURE 2.

WAXS profiles of selected samples.

The diffraction pattern intensities were normalized to the intensity of (110) reflection. It can be observed that the ratios of particular reflections differ from sample to sample, which is presumably caused by the varying extent of preferential orientation of the PCL crystallites with respect to the direction perpendicular to the film surface, which was caused by the solution casting process, as already observed for PCL‐containing solution‐cast films in our previous work [5]. The extent of preferential orientation can be assessed based on the comparison of ratios of particular reflections, for example, (110–200), in the respective WAXS profile with the corresponding ratio in the profile of PCL powder sample, in which the orientation of PCL crystallites is entirely random (isotropic material). Based on this comparison, it can be concluded that in the pure PCL film, there is some preferential orientation, whereas in the PCL/gel samples (PCL/B3, PCL/F3, PCL/P3), the orientation appears to be almost random. This may be due to the addition of a softer gelatin component, which presumably reduced the mechanical stress during the crystallization of PCL during the solution casting process. On the other hand, Figure 2 shows that MXene increased the extent of preferential orientation of PCL crystallites. It may have been caused by the steric effects of MXene nanosheets.

The purpose of the WAXS graph (Figure 2) is only to show the trends in the intensity ratios of particular PCL reflections and thus the trends in the preferential orientation of PCL crystallites to compare the powder PCL (in which the crystallite orientations are random), pure PCL film, PCL/gelatin films, and PCL/MXene/gelatin films. No clear trend of the PCL crystallite orientation was observed for varying amounts of gelatin in the particular sample series. Although the XRD reflection intensity ratios for different gelatin contents varied somewhat, the variation was within a range that allowed conclusions to be drawn about differences in preferential orientation between the three types of film samples (PCL, PCL/gelatin, and PCL/MXene/gelatin). Therefore, only WAXS curves of representative samples are shown.

The SAXS profiles of all the samples showed the typical PCL profile corresponding to the lamellar stack systems in PCL (Figure 3a). The samples with admixtures (i.e., with the different types of gelatin and the MXene particles) show little difference with respect to the pure PCL sample except for the small‐angle upturn in the low‐q region of the SAXS profiles. The small‐angle upturn indicates the presence of larger particles with sizes of at least 25 nm. The actual size of these objects cannot be determined due to the low‐q limit (q _min ~ 0.18 nm⁻¹) of the used instrument's q‐range.

FIGURE 3.

SAXS results: (a) SAXS profiles of selected film samples; (b) SAXS profiles of PCL/MX/B1 film, pure PCL film, and the difference profile; (c) 1D correlation function of PCL lamellar stacks calculated from SAXS profile of pure PCL film sample.

Figure 3b shows the SAXS profiles of the PCL/MX/B1 film sample (as an example), the PCL film sample, and the difference profile (PCL/MX/B1 minus the PCL), showing the profile corresponding to the mentioned larger objects alone. The profile due to these objects corresponds quite well to Porod's law (I ~ q ⁻⁴) plotted in the graph (the line with a slope of −4 in the shown double logarithmic plot). This fact indicates that the small‐angle upturn corresponds to the surface scattering (i.e., the high‐q part of the scattering) of homogeneous particles with a sharp interface, having their SAXS knee out of the range of the instrument (it would be found at q < q _min).

A comparison of the profiles of all the selected samples in Figure 3a shows that the small‐angle upturn is independent of the presence or absence of MXene. Thus, it cannot be assigned to the MXene, but to the gelatin particles in the PCL matrix, which were observed by electron microscopy (see Section 3.4). It is expected that MXene also contributes to the small‐angle upturn due to the dimensions of MXene particles in the range of tens of nanometers; however, as our comparison shows, their contribution to small‐angle scattering is minor compared to that of the present polymeric particles. This can be explained by the low weight percentage of MXenes in the sample (0.5 wt%) together with the fact shown by SEM and TEM (Section 3.4) that a substantial part of MXenes was present in the form of large agglomerates. If the MXene particles were well dispersed into nanosheets, then their contribution to the small‐angle scattering intensity would be substantial despite their low content of 0.5 wt% in the samples.

The long period of the PCL lamellar stacks was determined from the position of the first maximum of the 1D correlation function, Γ₁(z), of the lamellar structures [6] shown in Figure 3c, using the procedure described in [5], which resulted in 12.8 nm. Besides, the PCL long period was also determined from the first maximum of the Lorentz‐corrected SAXS profile as described in [5], which provided a value of 13.9 nm.

3.3. Mechanical Properties

The mechanical properties of films depend on many parameters, such as the type of present polymers, quantitative composition, type of dopants, particle size, and adhesion between minority polymer particles and dopant particles to the polymer matrix [110, 111, 112]. Sometimes, the interactions may contribute to the enhancement of mechanical properties or follow a reverse trend, leading to a decline in their properties. Hence, it is difficult to predict the exact properties of the material, especially when a natural polymer such as gelatin (obtained from fish, porcine, and bovine skin) is involved in designing the matrix.

Tensile ultimate strength, elongation at break, and tensile toughness were evaluated using static tensile tests. For each sample at least five stress–strain curves were measured and the evaluated parameters are thus averages from at least five values. Standard deviations were calculated, too. Representative stress–strain curves of pure PCL, PCL/B2, and PCL/MX/B2 composite samples are presented in Figure 4a. Figure 4b–d show the dependencies of the tensile ultimate strength, the elongations at break, and the tensile toughness, respectively, on the gelatin content in the investigated wt% range for all sample series.

FIGURE 4.

Results of static tensile tests: (a) representative stress–strain curves of PCL, PCL/B2 and PCL/MX/B2 samples; (b–d) dependence of ultimate strength, elongation at break, and toughness on the gelatin content in the samples, respectively.

The investigated parameters showed a large variance in general; however, some conclusions can be made based on the obtained results. The tensile experiments showed that the presence of gelatin led to a clear decrease in all the evaluated tensile parameters compared to the pure PCL film. The ultimate strength of pure PCL film was 11.8 ± 1.7 MPa, its elongation at break resulted in 288% ± 133%, and the tensile toughness was 30.8 ± 17.4 MPa. The reduction in the ultimate strength of PCL/gel samples compared to pure PCL was expected as the gelatin component is soft and amorphous as opposed to PCL. Further, it can be stated that the incorporation of such a relatively small content of MXene as 0.5 wt% altered the tensile properties of the films with the lower gelatin contents (12–15 wt%). The MXene‐containing films in this range of gelatin content showed a visible increase in all three investigated tensile parameters. The increase could be attributed to the compact structure that is created by the filling of MXene particles in the PCL/Gel matrix and to changes in the crystalline nature [70]. From WAXS profiles (Section 3.2), it is obvious that MXene changed the orientation of PCL crystallites, and this may have contributed to the enhancement of the mechanical properties. On the other hand, for the higher gelatin content of ~25%, no clear difference between the tensile parameters of samples with and without MXene was detected on average.

In [113] electrospun blends of PCL and gelatin were prepared and mechanically tested. The ultimate tensile strength of these samples was around 3 MPa for similar compositions as PCL/gel samples in this work. The corresponding elongation at break was in the range approximately 80%–100%. Thus, the values of these two mechanical parameters were comparable to the values obtained by us for PCL/gel samples; our values of ultimate strength of PCL/gel samples were somewhat higher (approximately 5–7 MPa) than those reported in the cited work, while the elongations at break were virtually the same.

3.4. Morphology From SEM and TEM Analyses

Morphology of the PCL/MX/Gel samples, some of which exhibited improved tensile strength (as explained in Section 3.3), was characterized by SEM microscopy (Figures 5 and 6), SEM/EDX microanalysis (Figure 7), and TEM analysis (Figure 8).

FIGURE 5.

SEM/SE micrographs showing fracture surfaces of all PCL/MX/Gel samples: (a–c) PCL/MX/bovine gelatin, (d–f) PCL/MX/fish gelatin, and (g–i) PCL/MX/porcine gelatin. The columns show, from left to right, the samples with the lowest (11.5 wt%), intermediate (15.5 wt%), and highest (24.5 wt%) concentrations of gelatin. The original magnification of all micrographs was ×2000.

FIGURE 6.

SEM/SE micrographs showing the surfaces of selected PCL/MX/Gel samples: (a) PCL/MX/bovine gelatin, (b) PCL/MX/fish gelatin, and (c) PCL/MX/porcine gelatin.

FIGURE 7.

Typical SEM/EDX spectra from the top (a) and bottom (b) part of a PCL/MX/gel samples. The insets show specific energy regions where weak titanium peaks should be observed. The Pt peaks originate from the Pt coating of the samples for SEM observations.

FIGURE 8.

Typical TEM/BF micrographs of PCL/MX/Gel composites showing (a) well‐dispersed MXene nanosheets, (b) small MXene agglomerates, and (c) very large, microscale MXene agglomerates.

Figure 5 shows the typical SEM/SE micrographs of cross‐sectional fracture surfaces (the SE abbreviation indicates that the micrographs were obtained with a secondary electron detector). The SEM/SE micrographs display mostly topographic contrast [91, 114]. Consequently, the SEM/SE images of polymer blends reveal the system's phase morphology. In our case we observed quite a coarse morphology of gelatin particles (smooth, rounded objects) in the PCL matrix (background with sharp fracture lines). As documented in Figure 5, the structure coarseness decreased in the order of PCL/MX/fish gelatin > PCL/MX/bovine gelatin > PCL/MX/porcine gelatin. The SEM/SE imaging was employed also in fast verification of the morphology of PCL/Gel samples (i.e., the samples without MXene) and the micrographs of PCL/MX/Gel and PCL/Gel samples were similar. In other words, the morphology of the systems was determined mostly by gelatin type and concentration, while the impact of MXene on the final system morphology was low.

Moreover, the SEM results suggested that at least some MXene nanoparticles formed agglomerates on a micrometer scale, which were observable as bright white spots in SEM/SE micrographs. The concentration of the agglomerates was independent of gelatin concentration, but it seemed to increase toward the bottom surface of the samples. This was further confirmed by SEM/EDX microanalysis and TEM/BF imaging, as discussed below.

On the other hand, all PCL/MX/Gel samples with the three different gelatin types showed approximately the same surface roughness, which is demonstrated in Figure 6 that shows SEM/SE surface micrographs of selected PCL/MX/Gel samples. The spherulitic structure of PCL is visible in these micrographs.

Figure 7 displays typical SEM/EDX spectra of a PCL/MX/Gel sample, which were obtained from the top surface (Figure 7a) and bottom surface (Figure 7b). These spectra were similar for all types of gelatin. The MXene nanoparticles (Ti₂C), which were added to the PCL/MX/Gel composites in this work at a concentration of 0.5 wt%, should exhibit additional weak titanium peaks in EDX spectra. The expected low intensity of the titanium peaks is related to the very low concentration of Ti in the prepared composites. The titanium peaks (TiKα at ~0.4 keV and TiLα at ~4.5 keV) were observed only on the bottom surfaces of the samples (compare Figure 7a with Figure 7b). This indicated that a substantial part of MXene nanofiller formed agglomerates that tended to sediment at the bottom of the samples.

Figure 8 is a collection of typical TEM/BF micrographs (BF stands for the classical bright field imaging) of the investigated PCL/MX/Gel composites. Part of the MXene nanofiller was quite well dispersed in the polymer matrix (Figure 8a). Still, a significant portion of the filler particles formed nanometer scale (Figure 8b) or even micrometer scale agglomerates (Figure 8c). The larger agglomerates tended to sediment during the sample preparation, as confirmed by SEM/EDX results (see Figure 7 and the discussion in the previous paragraph). The fact that the dark objects in TEM/BF were MXene agglomerates was confirmed by two observations: (i) in TEM/BF imaging, the MXene platelets, having a higher average atomic number, appeared darker than the polymer matrix with a lower average atomic number and (ii) the high‐resolution TEM/BF micrographs revealed a layered structure of the agglomerates, which is characteristic of MXene materials [115].

3.5. Cell‐Material Interaction

The cell culture experiments proved that the colonization of the composite materials with SAOS‐2 cells is influenced by the presence of MXene in the material and the type and concentration of gelatin in the composite. The fluorescence staining and microscopy of SAOS‐2 cells (Figure 9) show that the confluent layer of cells can be observed after 7 days of cultivation on all samples containing MXene and porcine gelatin at all concentrations. In contrast, the cells on samples containing MXene and bovine or fish gelatin were subconfluent and less homogeneously distributed, and the number of cells decreased with increasing gelatin concentration. The samples with higher gelatin content provided less mechanical support to the growing cells. Similar results to those obtained for samples with MXene and bovine or fish gelatin were obtained for samples with porcine gelatin not reinforced with MXene, where the least confluent layer of cells is formed on the samples with the highest gelatin concentration. Taken together, we can observe the negative effect of higher gelatin concentration in the composite and the improving effect of MXene in the samples with porcine gelatin but not in the samples with bovine or fish gelatin, probably due to their weaker mechanical properties.

FIGURE 9.

Fluorescence staining of SAOS‐2 cells after 7 days of cultivation—blue are cell nuclei stained by DAPI, and red are actin filaments. MX are samples containing 0.5 wt% of MXene, B—samples containing bovine gelatin, F—samples containing fish gelatin, P—samples containing porcine gelatin, PCL—control poly‐ε‐caprolactone samples without gelatin and MXene. PS—control tissue culture polystyrene. The representative images for this figure were selected from 15 images of the same sample type.

The measured metabolic activity (Figure 10) of the SAOS‐2 cells shows a significant decrease in metabolic activity on PCL control and studied composites on Days 3 and 7 compared to the tissue culture PS [116]. The reasons can be the fact that the basic PCL matrix is hydrophobic compared to the PS and the initial adhesion can be relatively slow, which in turn can influence the speed of proliferation in the later stages of the experiment. The other possible and perhaps more interesting cause of decrease in metabolic activity could be the switching of the cell metabolism toward differentiation [117]. This would have to be confirmed by other methods that have not yet been performed. Another possible explanation could be an unknown interaction of the material with the MTS test chemicals. This would probably have to be in a form of absorption of the MTS chemicals, thus lowering the absorbance in the particular wavelength. This would express itself as a coloration of the samples which was not observed. If we compare the composites to the pure PCL control it confirms the microscopic observations. Although the results are not statistically significant, we can observe certain tendencies that are present in the microscopy pictures. We can observe a decrease in the metabolic activity of the SAOS‐2 cells in higher concentrations of all types of gelatin on the 7th day of cultivation. We can also observe higher metabolic activity of SAOS‐2 cells on samples with porcine gelatin compared to bovine and fish gelatin. On the other hand, we can also see that on the first day of cultivation, the metabolic activity of SAOS‐2 cells is higher in the samples with bovine and fish gelatin. This suggests that cell adhesion is easier or faster on these two types of gelatin than on the porcine gelatin, which in turn provides a more suitable environment for cell proliferation. We can also observe a slight increase in the metabolic activity of the SAOS‐2 cells in the samples with the intermediate concentration (15 wt%) of fish or porcine gelatin compared to the samples with the lowest gelatin concentration. This can be observed in the microscopic images, especially in the samples with porcine gelatin and without MXene.

Automated segmentation of DAPI‐stained nuclei enabled comparison of cell numbers among the tested substrates (Figure 11). The nuclei counts partially corresponded to the trends observed in the MTS assay, although not all differences reached statistical significance. Variability between fields was relatively high, reflecting the intrinsic surface heterogeneity of the fibrous substrates. Interestingly, substrates with higher gelatin content exhibited a tendency toward reduced nuclei numbers, which was consistent with the lower metabolic activity measured by MTS. It should be noted that metabolic assays and cell counts provide complementary information. While nuclei quantification reflects cell number and surface coverage, MTS readout depends on the overall metabolic state of the cell population, which may decrease at higher cell densities due to reduced spreading and contact inhibition. Therefore, direct proportionality between these two parameters cannot be expected.

The improvement of tensile properties upon MXene incorporation can be attributed to enhanced interfacial interactions between the filler and the PCL matrix, leading to more efficient stress transfer within the composite structure. Beyond mechanical reinforcement, we also expected changes in surface‐related parameters that are relevant for cell–material interactions. MXenes are known to exhibit high surface energy, hydrophilicity, and the presence of surface functional groups (–OH, –O, –F), which may alter protein adsorption dynamics and subsequent cell adhesion. In addition, their electrical conductivity has been reported to modulate cellular behavior in certain systems, particularly, in electrically responsive tissues. However, in the present study, the biological response is more likely governed by surface heterogeneity in composition and topography, and also by the source of the gelatin used to fabricate the composite rather than by bulk mechanical properties alone.

The relationship between mechanical stiffness and cell behavior is complex. While moderate increases in substrate stiffness can promote cell spreading and cytoskeletal organization, excessive heterogeneity or coarse morphology may counteract this effect. Therefore, the mechanical enhancement observed in the composite films does not necessarily translate directly into improved metabolic activity.

Differences among gelatin sources may further contribute to the observed biological variability. Gelatin derived from different species differs in amino acid composition, bloom strength, isoelectric point, and molecular weight distribution. These parameters influence surface charge, hydration behavior, and protein–cell interactions. The tendency of higher gelatin content to reduce cell number under the present conditions may reflect changes in surface stability, swelling behavior, or protein adsorption rather than a purely compositional effect.

Our findings suggest that, in solution‐cast PCL/gelatin films, the type and concentration of gelatin have a more pronounced impact on cell behavior than MXene incorporation. While MXene effectively enhanced mechanical strength, its biological contribution under the present conditions appears limited.

4. Conclusions

In this study, a series of PCL/MXene/gelatin composite films were prepared and systematically characterized in terms of their morphological, mechanical and biological properties to assess their suitability for biomedical applications. The biological analysis revealed a decrease in metabolic activity of cells compared to PS controls; however, relatively high cell confluence was maintained for samples containing up to 14.5 wt% gelatin. The biocompatibility of the samples was influenced by both the gelatin type and the presence of MXenes. Surprisingly, a higher concentration of gelatin hindered cell proliferation on the composite, regardless of the gelatin type. For the porcine gelatin series, the inclusion of MXene did not result in a significant difference in cell proliferation or metabolic activity. The incorporation of MXene enhanced the tensile properties of the films for composites with lower gelatin contents; the ultimate strength of the samples ranged from approximately 6–10 MPa. Morphological analysis revealed a coarse microstructure of PCL/MX/Gel composites, with particle sizes in the order of magnitude from units to tens of μm. The degree of structure coarseness decreased in the following order: PCL/MX/fish gelatin > PCL/MX/bovine gelatin > PCL/MX/porcine gelatin.

Overall, the investigated films, in particular, the porcine‐based samples with MXene (PCL/MX/P1‐3), demonstrated strong potential for biomedical applications due to the favorable combination of mechanical performance and biological response. These materials may be suitable, for example, as scaffolds for hard tissue engineering or as biomaterials to modulate cell‐material interactions. Nevertheless, further work is required to advance these systems toward practical use. Future studies should explore alternate MXene types with reduced agglomeration to improve dispersion in the polymer matrix, as well as fabrication methods beyond solution casting to achieve better component intermixing and finer microstructural features.

Funding

This work was supported by the European Regional Development Fund (CZ.02.01.01/00/22_008/0004596), Ministry of Education, Youth and Sports of the Czech Republic (MSMT‐21927/2023), Czech Science Foundation (Grant No. 26‐21031S), and Czech Academy of Sciences (AP2202).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Illustrative images of the efficiency of Cellpose nuclei segmentation. (A) Original grayscale image of cell nuclei. (B) Result of the Cellpose image segmentation process—colored mask showing individual objects/nuclei found by the Cellpose tool.

Data Availability Statement

The data described in the article are available at https://doi.org/10.5281/zenodo.13379919. We would appreciate it if other researchers could benefit from our results. This will foster discussions and collaboration among scientists worldwide.