4.1 Morphological characterization of COL-rGO-PDA hydrogels
The samples’ morphology was analyzed by scanning electron microscopy (SEM) and the micrographs are shown in Fig. 1A-I.
GO SEM image (Fig. 1A) reveals a structure of wrinkled and randomly aggregated layers, with the absence of flat surfaces and with orderly layer structures characteristic of graphite, which is a product of the oxidation. The SEM image of rGO (Fig. 1B) reveals a decrease in size and an increase in the number of wrinkled sheets with a more disordered structure than GO and a tendency to become entangled with each other, caused by van der Waal interactions [28].
The hydrogels’ morphology is observed in Fig. 1C to F, where an increasing rGO amount produced a significant growth in surface irregularity, which could induce a larger contact area between cells and the material. In Fig. 1G and 1I, the presence of collagen fibers was noticed which increases the porosity of the hydrogel and thus the fluid adsorption.
4.2 Chemical characterization of COL-rGO-PDA hydrogels
FTIR spectroscopy was used to identify the chemical structure of GO, rGO, DA, glycerol, COL, and COL-rGO-PDA hydrogels. The GO spectrum (Fig. 1J) shows the bands associated with the oxygenated groups, the vibration of the surface hydroxyl groups (C − OH) being visible, in addition to the absorbed water molecules in a broad band at 3335 cm− 1.The bands related to C = O (− COOH) vibrations are visible at 1730 cm− 1 and the stretching modes of the epoxy (C − O−C) and alkoxy (C − O) groups are observed at 1194 and 1050 cm− 1, respectively. The band at 1635 cm− 1 can be attributed to the aromatic double bond (C = C) vibrations of the sp2 carbon skeletal lattice and to the C − H bending vibration frequencies of the carbon lattice [16, 29].
DA spectrum has bands at 1619 and 1500 cm− 1 corresponding to the C = C stretching vibration of the benzene ring and the − OH stretching vibration of the DA diols [30]. In the rGO spectrum (Fig. 1J), the bands associated with the stretching vibrations of the C = O (amide I) groups appear at 1645 cm− 1 and the N − H bending and C − H stretching vibrations (amide II) at 1546 cm− 1. A decrease in the intensity of the bands associated with the epoxy and alkoxy groups is observed at 1194 and 1050 cm− 1, respectively. These elements demonstrate the successful reduction of GO with DA and, simultaneously, the rGO functionalization by DA amino groups.
The FTIR spectrum of COL-rGO-PDA (Fig. 1K) shows the characteristic bands corresponding to amides I, II and III at 1653, 1545 and 1239 cm− 1, respectively [31]. A light change of amide I to a higher frequency (1645 to 1653 cm− 1) can be observed, since the C = O stretching vibrations of rGO and amide I of COL occur in this band. The peak at 1410 cm− 1 is attributed to the C − OH bonds of the catechol groups from DA, indicating that DA has been successfully introduced into the COL hydrogel [32]. A peak around 1350 cm− 1 is observed in all hydrogels, which can be attributed to a stretching of the indole ring, supporting the formation of 5,6-dihydroxyindole/5,6-indolequinone units that are characteristic of COL-DA hydrogels [32].
The broad band between 3200–3600 cm− 1 is caused by the -OH groups stretching of the collagen [33]. Moreover, the peak at 3350 cm− 1 can be specifically attributed to the presence of hydrogen bonds formed between intermolecular water, glycerol, and collagen amide A. These hydrogen bonds are associated with the stretching vibrations of the N-H groups [32, 34].
It was determined that varying concentrations of rGO have no significant impact on the absorbance of amide A. This finding suggests that the water molecules bound to the collagen were not lost, enabling the formation of new bonds with different molecules. The characteristic bands of glycerol can be observed at 1040 and 1110 cm− 1, indicating the stretching of C − O bonds in primary and secondary alcohols, respectively [35]. The spectral region between 2800–3000 cm− 1 corresponds to the bands linked to C − H aliphatic bonds. These bands can be attributed to the presence of glycerol lipids found in the hydrogel [34].
An increase in the intensity associated with the amide bands (I, II and III) and a disappearance of the peak at 1450 cm− 1 were observed when increasing the amount of rGO to the hydrogel, which can be assigned to the bending vibrations of CH2 [36, 37]. These differences in intensity suggest the formation of new hydrogen bonds among the hydroxyl, amino and carboxilc groups [38].
The effect of the components grafted on the hydrogel on the stability of collagen can be evaluated through the amide I profile, where the α triple helix band is observed at 1650 cm− 1 [38], and the formation of the β sheet was not perceived [39]. COL hydrogel’s stability was calculated through the relation between the amide III and the aliphatic band of bonds C-H (1456 cm− 1) [40]. This relation was higher than 1 for all synthetized materials, showing that rGO grafting did not affect the α triple helix structure.
The cross-linking degree is a key aspect that must be controlled in hydrogels since it is directly related to their pore size and absorption capacity [41]. The ratio between amide A/amide I was used as a collagen reticulation measure, revealing that increased rGO favors the cross-linking density.
The hydrogels’ crystallinity was evaluated through XRD patterns (Fig. 1L). The XRD pattern of the GO revealed a very sharp diffraction peak at 2θ = 8.68° and an interlayer distance of 9.86 Å; these elements demonstrate the complete oxidation of graphite and the intercalation of oxide functional groups on the carbon basal plane, such as epoxy, hydroxyl, carbonyl, and carboxyl groups [42, 43]. After the GO reaction with DA, a wide pick was obtained (18° to 28°) with a maximum intensity at 2θ = 22.5, reducing the interplanar distance to 3.96 Å. These findings can be attributed to the elimination of oxygenated functional groups and the \(\varvec{\pi }-\varvec{\pi }\)stacking among the hexagonal cells of graphene and the DA aromatic ring, showing a favorable GO reduction [16, 17].
COL hydrogel presented three characteristics peaks within its structure: the first one at 6° shows the intermolecular side packing among the collagen molecular chains; then, at 20° there is the diffuse association of collagen; and the third peak, at 33°, is associated with the residual aminoacid of the triple helix [44–46]. Moreover, the diffractograms of the samples revealed a scattered distribution of collagen fibers, with a distinct peak of rGO at 6°. The intensity of the peak increases as the concentration of rGO rises. The peak centered around 2θ = 22° shows a proportional increment in intensity corresponding to the amount of rGO. This can be attributed to the masking effect on the collagen spectrum and the emergence of the distinctive rGO peak. The disappearance of the peak at 33° may be attributed to interference caused by the presence of a highly crystalline structure, like rGO.
The surface compositions of GO, rGO, pure collagen, COL hydrogel and COL/rGO50 were studied by XPS (Fig. 1M). It appears that carbon (C1s) and oxygen (O1s) were the predominant components in all the systems, with binding energies of 284.6 and 531.9 eV, respectively. Furthermore, nitrogen (N1s) was detected in the rGO, pure collagen, COL and COL/rGO50, with a binding energy of 399.7 eV.
The impact of DA reduction was assessed by comparing the oxygenated group content in GO and rGO, through the calculation of the areas under the C1s and O1s peaks. The C/O ratio of GO was found to be lower (2.37/1) as compared to rGO (4.96/1), indicating the successful reduction of oxygenated functional groups in GO. Similar values for the C/O ratio have been obtained in other studies, with GO and rGO exhibiting ratios of 2.75/1 and 5.2/1 respectively [19]. The C1s deconvolution spectra of GO and rGO (Fig. 1N and 1O) enabled the identification of the oxygenated groups that were most extensively eliminated. The positions of the GO peaks were 284.6 eV (C = C sp2-hybridized carbon, and C-C sp3-hybridized carbon), 286.5 eV (C-OH), 286.9 eV (C-O-C) and 288.2 eV (O-C = O) [19]. In the deconvolution of the rGO peaks, changes can be noticed. The C-C bond maintains its position at 284.6 eV, while the C-N bond is observed at 285.9 eV. The presence of the C-N bond can be attributed to the combination of the DA amino group and the carbon atoms in rGO. These findings were consistent with the observations made in the FTIR analysis. Additionally, the C-OH and C-O-C bonds are now centered at 266.7 and 287.1 eV, respectively. The observed increase in binding energies can be attributed to the removal of oxygenated functional groups present in GO, which had a shielding effect on the valence electrons. This indicates a higher concentration of carbon atoms with more robust bonds, as depicted in Fig. 1S. Following the reduction process, there was an increase in the percentage of C-OH bonds from 9.8–23.4% on rGO. This increase can be attributed to the presence of hydroxyl groups associated with the aromatic ring of DA.
The deconvolution of C1s in pure collagen (Fig. 1P) revealed distinct peaks located at 284.6, 285.9, and 288.2 eV. These peaks can be associated with C-C, C-N, and O-C = O bonds, respectively [23]. The combination of DA and glycerol (Fig. 1Q), along with the catalytic action of HRP/H2O2 within the hydrogel sample, leads to the formation of C-OH bonds and a rise in O-C = O bonds. The presence of C-OH bonds is primarily due to the DA catechol groups, rather than glycerol. This is because glycerol is typically absorbed into the hydrogel structure and does not remain on the surface [47]. The inclusion of rGO (Fig. 1R) favors the formation of C-N bonds, supporting the FTIR findings that suggest the creation of amide bonds with collagen. Furthermore, the potential partial oxidation of rGO by HRP/H2O2 in the hydrogel matrix is unlikely to be based on the observed decrease in the concentration of C-OH and O-C = O bonds. Therefore, the observed increase in the intensity of the peak with lower binding energy indicates a successful interaction of collagen in the synthesis process [48].
4.3 Thermal Stability of COL-rGO-PDA hydrogels
The thermal stability of the hydrogel samples was analyzed by thermogravimetric analysis (TGA) and differential thermogravimetric analysis (DTG) (Fig. 2D, Fig. S1 and Table 1).
Table 1
Thermal analysis data for the hydrogel samples
Sample | Th 1 (°C) | Ti 2 (°C) | Tmax 3 (°C) | Tb 4 (°C) | M6005 (%) |
COL hydrogel | 83 | 225 | 286 | 337 | 7.6 |
COL/rGO25 | 92 | 211 | 280 | 335 | 13.6 |
COL/rGO50 | 100 | 193 | 275 | 331 | 18.7 |
COL/rGO100 | 115 | 184 | 270 | 317 | 24.2 |
1 Th: the extrapolated onset temperature of the water loss peak, 2 Ti: the extrapolated onset temperature of the thermal decomposition peak, 3 Tmax: the temperature at the maximum weight loss rate, 4 Tb: the extrapolated final temperature of the DTG curve. 5 M600: the residual carbon content at 600°C.
In all synthesized samples, the thermograms displayed two distinct decomposition processes. The first occurred between 83–115°C, which indicated the evaporation of water. The second decomposition step took place between 184–337°C, being related to the deterioration of the collagen fibers. The increase in rGO concentration raised the degradation temperature associated with moisture loss. The first inflection in the curve refers to the water absorbed, indicating that the polymers are sensitive to moisture absorption, which may increase the swelling capacity of hydrogels.
Specifically, the collagen thermal degradation started at 225°C, with a main weight loss at 286°C, a value lower than the decomposition range reported for native collagen (300–400°C) [49]; this difference could be caused by the loss of low molecular weight proteins and the presence of glycerol in the hydrogel. Other studies working with collagen film and glycerol reported a degradation temperature of 223°C [40], values even lower to those obtained here. This thermal strengthening could be caused by an increase in the cross-linking density generated by the HRP/H2O2 and the polymerization of DA. Studies have shown that the collagen triple helices are stabilized by numerous covalent bonds generated by HRP, which increases the denaturation temperature [50, 51].
The COL hydrogel decomposition temperature was 225°C, and the rGO incorporation lowered it to 211, 193 and 184°C for COL/rGO25, COL/rGO50 and COL/rGO100, respectively. These changes indicate that enzymatic and chemical cross-linking between the polymers occurred in the hydrogel matrix. The observed decrease in heat released during degradation can be attributed to the high thermal conductivity of rGO. In fact, the incorporation of rGO in the hydrogels facilitates the dissipation of heat produced during the degradation process. Moreover, the residual carbon content increases with increasing rGO concentration. Thus, the thermograms suggest that the rGO was distributed in the hydrogel and that the obtained biocomposite hydrogels were thermally stable under the working conditions.
4.4 Swelling capacity, penetration mechanism and wettability
The swelling capacity of the materials was determined in PBS at different pH values in order to assess their anionic or cationic behavior. The absorption curves (Fig. 2A to C) reached the equilibrium state after 15 min, for all samples, with different maximum values.
The hydrogel sample with highest absorption capacity was COL/rGO25, reaching 215, 151 and 88 g/g at pH values of 2, 4 and 6, respectively. The swelling capacity of the materials increases considerably, when the medium pH decreases, evidencing the anionic characteristics of the materials, with a predominance of protonatable functional groups. Moreover, the hydrogel´s absorption capacity decreased with the increasing rGO concentration (from 25 to 100 g). The observed phenomenon can be attributed to the preservation of the hydrogel's original morphology by the rGO and the occupation of the hydrogel's pores. The kinetic parameters obtained after adjusting the Fick Law to the experimental data are presented in Table S1. Most of the hydrogels had an exponent n of less than 0.5, indicating Fickian diffusion. Values over 0.5 indicate the presence of abnormal transport, where an overlap of molecular diffusion and stress relaxation may occur during swelling [52]. A decrease in the values of n can be observed at pH = 4 and pH = 6 with increasing rGO concentration, which is attributed to a protective effect of rGO on collagen fibers and a lower probability of their relaxation. This wide range of values indicates the different swelling mechanisms of the synthesized hydrogels. Additionally, the differences between the k values are due to a different water intake rate in the synthesized materials. These elements could suggest a heterogeneity in the pore size as a function of the rGO concentration.
Contact angles of the rGO, GO, COL hydrogel, and COL-PDA-rGO hydrogels were used as a measure of the hydrogel’s wettability (Table 2).
Table 2
Contact angle, conductivity and surface charge
Sample | Contact angle (º) | Conductivity (mS/m) | Surface charge (mV) |
GO | 61.8 ± 1.4 | 121.4 ± 12.3 | -32.8 ± 2.3 |
rGO | 84.5 ± 2.0 | 232.6 ± 10.6 | -70.2 ± 5.4 |
COL hydrogel | 50.6 ± 1.3 a | 20.27 ± 0.25 a | 4.5 ± 0.8 a |
COL/rGO25 | 54.1 ± 1.6 a | 24.61 ± 0.03 b | -46.8 ± 1.1 b |
COL/rGO50 | 73.6 ± 2.6 b | 29.72 ± 0.61 c | -65.2 ± 2.2 c |
COL/rGO100 | 76.4 ± 2.8 b | 39.57 ± 0.19 d | -79.0 ± 3.1 d |
The contact angles of GO and rGO were 61.8 and 84.5º, respectively. These values were similar to those obtained in other studies [41, 53]. This increase may be caused by the elimination of functional groups present in the GO during the reduction process, making the surface more hydrophobic. The contact angle of COL hydrogel was 49.8º, it being more hydrophilic than both GO and rGO. This difference in the sample hydrophilicity can be attributed to the unique behavior of collagen's triple helices, which allows the absorption and retention of water [54].
All the hydrogel samples showed a contact angle of less than 90º. The incorporation of DA, glycerol and HRP/H2O2 into the collagen increased the contact angle from 49.8 to 50.6º for the COL sample, caused probably by the formation of bonds among the components, increasing the cross-linking collagen density. As expected, increasing the rGO concentration enhances the hydrophobicity of the material (Table 2). This can be attributed to the presence of a limited number of sites that can form hydrogen bonds with water. These transformations in the hydrogel hydrophobicity indicate changes in the chemical surface and are the consequence of a successful chemical/enzymatic reaction.
4.5 Conductivity and Surface Charge
The method used to determine the electrical resistance of the materials was the 4-point method. Table 2 shows the conductivity values of the materials determined from the resistance measured. The values fluctuated between 20.02 and 39.76 mS/m, where the conductivity of hydrogels is directly correlated with rGO concentrations, which can be attributed to the creation of an additional electric field [55]. It is important to note that the electrical conductivity of human skin falls within the range of 10− 2 to 260 mS/m [56]. Scientific studies have shown that using conductive dressings with similar conductivity values to the skin enhances the effective transmission of electrical signals and directs various cellular mediators, such as platelets, inflammatory cells, cytokines, growth factors, and matrix metalloproteinases, towards the center of tissue injuries, thus facilitating tissue repair [57–60].
The conductivity values of the COL hydrogel were found to be significantly higher than those achieved by Norahan et al. [61] in their study on COL/rGO hydrogels with a ratio of 1:0.08 who reported conductivity values of only 0.12mS/m. The highest conductivity values obtained may be caused by the presence of nitrogenous groups in the collagen structure, which in the presence of water lose an electron, acquiring the ability to accept electrons, thus facilitating their movement [62]. The glycerol influence was also detected; this compound has a large number of hydroxyl groups that increase water retention and reduce the intermolecular forces between the polymer chains, improving electrical and mechanical properties [63]. Finally, the high humidity of the hydrogels was a parameter that could also favor their conductivity, since the auto-ionization of the water creates a constant ionic flow in the hydrogel matrices.
The surface charge of the materials was determined by DLS. The GO surface charge was − 32.8 ± 2.3 mV, and after DA reduction, it reached a value of -70.2 ± 5.4 mV (Table 2). The negative values observed for GO can be attributed to the ionization of carboxyl groups, indicating that the oxidation reaction in the synthesis process was appropriately washed [53]. The removal of oxygenated functional groups during reduction restores the electronic configuration of the graphene structure and leads to a higher density of free electrons available for charge transfer [64]. The surface charge of the hydrogels varied from 4.5 to -79.0 mV, suggesting that the presence of glycerol, DA, and rGO alters the surface properties (Table 2). The incorporation of DA and glycerol into the matrix caused a decrease in surface charge from 4.5 ± 0.8 to 9.2 ± 0.4 mV. This change can be attributed to the neutralization of carboxyl and amino groups, as a result of their interaction with these compounds. The slight variation observed indicates that these compounds did not affect the helical structure of collagen. The rGO inclusion into the matrix resulted in a noticeable reduction in the surface charge, which could be attributed to the deprotonation of the collagen's carboxyl groups upon interaction with rGO, as well as the ability of rGO to absorb negatively charged species into the solution [64].
An additional study examined the impact of pH on the surface charge of dissolved collagen (Table S2). When the collagen was dissolved in HCl (0.05 M, pH = 4), the zeta potential measurements showed values of 12.3 ± 0.4 mV. This change in surface charge can be explained by the protonation of carboxyl groups, specifically those in glutamic acid and aspartic acid, within the collagen structure [65]. As the pH increased, a noticeable decrease in surface charge was observed, reaching a value of -13.5 ± 0.2 mV at pH = 9, which can be attributed to the deprotonation of carboxyl groups [65, 66].
4.6 Biodegradability and Cytotoxicity Assays
The in vitro biodegradability of the hydrogels was evaluated using collagenase I, as shown in Fig. 2E. During the first three hours, a substantial increase in the degradability rate was observed, being similar for all samples. This is due to the high concentration of collagenase I in the medium. Moreover, by increasing the rGO concentration, a higher resistance to biodegradation was observed, probably due to bigger interactions between collagen and rGO that strengthen the structure of the hydrogel matrix. Between 18 and 24 hours of immersion, a decrease in the degradation rate was observed, (1.1, 1.1, 0.9 and 0.2% for COL hydrogel, COL/rGO25, COL/rGO50 and COL/rGO100, respectively). It can be assumed that the interaction between rGO and the surrounding polymer chains created physical barriers that restricted the access of collagenase type I to the hydrogel network, thereby effectively reducing the degradation rate. The absence of biodegradation supports the conclusion that their stability in this medium is primarily due to their synthetic composition. This finding is crucial for the development of hydrogels for various biomedical applications where long-term stability and minimal degradation are desired.
In other studies where collagen-based hydrogels were synthesized, degradation rates of 50–60% were obtained using the same concentration of peroxidase [67]. Maximum degradation obtained for COL-rGO-PDA hydrogels was 25%, so it can be concluded that the polymerization of dopamine and the enzymatic cross-linking used also provide a protective effect against biodegradation activity.
Wound dressings should possess different qualities, such as: biocompatibility, non-toxicity, and the ability to facilitate cell migration [41]. The biocompatibility of all synthesized hydrogels exhibited a rate higher than 80% (Fig. 2F). The viability of the pure collagen obtained was higher compared to that of the control (DMEN supplemented with 10% FBS and 1% antibiotic) due to its natural presence in human body tissues and excellent compatibility with the biological environment [68]. Moreover, it possesses a stable chemical structure that contains arginine-glycine-aspartic acid sites, which favor cell adhesion and proliferation [69, 70]. rGO had a cell viability rate of 98.5 ± 4.3%. Previous research has indicated that the removal of oxygenated functional groups enhances cell biocompatibility, and the reduction achieved with DA does not pose a risk, in comparison with other reduction techniques that negatively affect cell viability [16, 71].
The rGO incorporation into the matrix had no toxic impact, and, in fact, promoted the survival of cells. Previous research has demonstrated that the water retention capacity of the hydrogel is heightened through the incorporation of rGO leading to enhanced hydration and support for cellular interactions and viability [72]. Moreover, the roughness of the material has been found to facilitate interactions between functional groups and cells, potentially activating molecules responsible for cell adhesion and proliferation [73], an effect that was also observed for this study.
4.7 In Vitro Wound Healing Assay (Scratch Test)
Fibroblasts play a crucial role in the tissue repair process, due to their ability to migrate and proliferate [74]. The ability of cells to migrate was evaluated by measuring the closure of a wound created in a fully confluent monolayer of fibroblast (Fig. 2G). The optical microscopy images (Fig. 2H) indicate the migration of fibroblast cells towards the scraping zone. It can be observed that this migration occurred after 24 hours of incubation with COL/rGO50 and COL/rGO100 samples, and after 48 hours for COL and COL/rGO25 samples. Subsequently, an accelerated migration of cells was observed, which was attributed to the elevated density of negative charges detected on the hydrogel's surface. Various studies have reported that biomaterials with negative charges promote the proliferation stage in the process of wound healing, facilitating the migration and proliferation of cells to effectively cover the injured area [75]. This observation is further supported by the wound closure rates determined in this study. In the first 24 hours, no significant differences in wound closure rates were observed (Fig. 2H). However, between 24 and 36 hours, all tested hydrogels exhibited a substantial decrease in the torn area. In fact, the rGO inclusion in the matrix had a positive impact on the cell migration process. Specifically, the COL/rGO100 material showed a closure rate of 84.5% after 48 hours of contact. The observed phenomenon can be associated with the increase in the hydrogel´s conductivity, enabling the regeneration of the skin's intrinsic electrical fields. These fields stimulate signals that direct the movement of epithelial cells towards the wound site, facilitating an ordered cell migration process that promotes efficient tissue healing [76–78].