Synthesis, Characterization, and In Vitro Evaluation of Lignin-Chitosan-Agarose Hydrogel Combined with Molecular Simulations for Potential Anti-Acne Patch Applications

1. Introduction

Acne vulgaris is a prevalent chronic inflammatory disorder that affects a large proportion of adolescents and adults worldwide. It is characterized by hyperkeratinization of hair follicles, increased sebum production, and colonization by Cutibacterium acnes (formerly Propionibacterium acnes), which contributes to inflammation and lesion formation [1,2]. The bacterial lipase enzyme secreted by C. acnes plays a pivotal role in hydrolyzing triglycerides in sebum into free fatty acids, exacerbating the inflammatory response and worsening acne symptoms [3,4].

Salicylic acid (SA), a beta-hydroxy acid, is widely used in topical acne treatments for its keratolytic, anti-inflammatory, and antimicrobial properties [5,6]. However, conventional formulations often suffer from rapid drug release and poor skin retention, which can reduce therapeutic efficacy and increase the risk of irritation [7,8]. In contrast, other site-specific delivery systems such as acne patches are designed to provide an initial burst release followed by sustained drug release over several hours, with most of the incorporated drug being released within 6–12 h to maintain therapeutic concentrations at the treatment site [9,10,11]. Therefore, there remains a need for short-term controlled-release systems capable of delivering SA in a similar therapeutically relevant release profile while improving efficacy and minimizing adverse skin reactions.

Hydrogels composed of natural polymers have gained significant attention as drug delivery platforms due to their biocompatibility, biodegradability, and tunable physicochemical properties [8,12]. Among them, lignin, chitosan, and agarose are promising candidates for hydrogel fabrication. Lignin is an abundant aromatic biopolymer derived from plant biomass with antioxidant and antimicrobial properties, making it attractive for biomedical applications [13,14]. Chitosan, a cationic polysaccharide obtained from chitin deacetylation, exhibits intrinsic antimicrobial activity and excellent biocompatibility, which can synergistically enhance the antibacterial effect of the hydrogel [15,16]. Agarose, a neutral polysaccharide extracted from seaweed, forms thermoreversible gels that contribute to hydrogel mechanical strength and stability [17,18].

Recent studies have demonstrated the successful incorporation of lignin into chitosan and agarose matrices to form composite hydrogels with improved mechanical properties and drug-delivery capabilities [19,20]. For instance, lignin-chitosan-agarose hydrogels have been shown to possess highly porous structures that facilitate drug loading and controlled release [21]. Furthermore, lignin-based hydrogels have been explored for their potential in antibacterial applications, including targeting C. acnes [22].

In addition to experimental characterization, computational approaches such as molecular docking and molecular dynamics simulations provide valuable insights into the molecular interactions between drugs and bacterial targets. The interaction of SA with C. acnes lipase, particularly subunit A of the enzyme (PDB ID: 6KHM), can reveal potential inhibitory mechanisms that contribute to its therapeutic effect [23,24].

In this study, we present an integrated experimental and computational approach to develop a lignin:chitosan:agarose hydrogel as a short-term sustained-release carrier for SA. The hydrogel composition was systematically optimized to identify the most suitable formulation for drug incorporation, followed by comprehensive physicochemical characterization, in vitro drug-release testing, and antibacterial evaluation. Furthermore, molecular docking and molecular dynamics simulations were performed to elucidate the interaction between SA and C. acnes lipase, providing mechanistic insight into its antibacterial action. This work provides a more comprehensive framework for the rational design of natural polymer-based anti-acne hydrogels than studies relying solely on experimental characterization.

2. Results and Discussion

2.1. HG-B Formula Optimization and Characterization

Lignin and chitosan are well-known biopolymers with excellent biocompatibility, thermostability, and durability characteristics, making them favorable for sustainable hydrogel development [25,26]. In this study, we optimized the hydrogel (HG-B) formula with various lignin-chitosan composition ratios (Table S1) to obtain the most highly swelled and structurally stable hydrogel. A lignin-chitosan hydrogel was prepared by a conventional blending and gelling method using agarose as a gelling agent. Agarose induces self-gelling owing to its numerous hydroxyl groups, which facilitate the formation of a helical structure [27] and trigger physical cross-linking via hydrogen bonding with lignin and chitosan, leading to 3D network formation [28]. The prepared HG-B formed a compact brown hydrogel, as depicted in Figure S1a.

Since the self-gelling process depends on the hydrogen-bonding site, hydrogel gelation time may be affected by lignin-chitosan compositions [29,30]. Thus, the gelation time of the various HG-B was investigated. As shown in Table 1, HG-B with a higher chitosan composition (HG-B3) gives the fastest gelation time. This phenomenon can be attributed to the presence of amine and hydroxyl groups in chitosan, which readily participate in hydrogen bonding. These intermolecular interactions facilitate the rapid formation of a continuous and stable hydrogel network [31]. In contrast, HG-B3, which has a higher lignin composition, exhibited the slowest gelation time. The rigid aromatic structure and relatively hydrophobic nature of lignin reduce the mobility of polymer chains within the system, thereby slowing the formation of the gel network [32]. On the other hand, HG-B2 exhibited a moderate gelation time. Such a gelation process allows sufficient time for polymer chains to rearrange and establish intermolecular interactions before the gel network is fully formed, resulting in a more stable and homogeneous hydrogel structure [33].

The swelling behavior of the hydrogel is a characteristic that affects its exudate absorption capacity, helps maintain a moist environment, and regulates controlled drug release [34]. HG-B2 showed rapid swelling rates (737.61%) and water retention within 1 h (Table 1). The hydrogel reached maximum swelling at 48 h (996.12%) with no scaffold disintegration or breakage, indicating excellent durability and stability (Figure S1b). On the contrary, HG-B1 disintegrated after 6 h of immersion, whereas HG-B3 exhibited inferior swelling performance, as indicated by its relatively low swelling ratio (374.37%). This indicates that a balanced lignin–chitosan ratio forms a polymeric network that enables uniform water diffusion while maintaining structural integrity, thereby enhancing hydrogel stability during prolonged immersion [35]. Based on these observations, HG-B2 serves as the optimal hydrogel compared to other hydrogels.

To investigate the chemical interactions and confirm the successful incorporation of lignin, chitosan, and agarose within the HG-B2 matrix, Fourier-transform infrared (FTIR) spectroscopy was employed (Figure 1a). The FTIR spectrum of the composite hydrogel displayed characteristic peaks corresponding to each component. Broad absorption bands around 3200–3400 cm^-1 were attributed to overlapping –OH and –NH stretching vibrations, suggesting the presence of extensive hydrogen bonding among the phenolic groups of lignin, the amino groups of chitosan, and the hydroxyl groups of agarose [36]. A peak appeared at 2939 cm^-1 attributed to the C–H stretching of lignin. Notably, the peak associated with the C=O stretching of lignin appeared at 1639 cm^-1, while aromatic C=C stretching vibrations were observed around 1563 cm^-1 [37]. The appearance of the spectrum between 1075 cm^-1 and 1022 cm^-1 is correlated with C–O stretching of chitosan [38]. Shifts and broadening of these peaks compared to the spectra of individual components suggested strong intermolecular interactions, particularly hydrogen bonding, which likely contributed to the structural stability of the hydrogel and drug encapsulation efficiency [39].

Scanning electron microscopy (SEM) images revealed a highly porous and interconnected three-dimensional network within the hydrogel (Figure 1b). The pore sizes ranged approximately from 10 to 50 µm, creating an architecture conducive to both drug loading and diffusion. Such porosity is critical for enabling the sustained release of SA by providing channels for the gradual diffusion of drug [40]. Furthermore, the porous architecture facilitates efficient absorption of wound exudate and helps maintain a moist wound environment, both of which are advantageous for topical acne treatment [41]. BET analysis was performed as complementary evidence of hydrogel porosity. The N₂ adsorption and desorption isotherms are depicted in Figure S2. From the isotherm result, the curve exhibited a type IV isotherm with an H3 hysteresis loop (IUPAC), indicating a mesoporous material [42]. BET analysis revealed that the HG-B2 possessed a specific surface area (S_BET) of 83.964 m²/g and a total pore volume of 0.107 cm³/g.

2.2. SA Drug Incorporation into Lignin-Chitosan Hydrogel (HG-SA)

Based on the previous observations, HG-B2 was selected as the optimal hydrogel formulation for subsequent SA incorporation to yield various HG-SA. SA was incorporated into the hydrogel by in situ loading during gelation (Figure 3a). Briefly, SA was mixed with lignin-chitosan according to the predetermined ratios presented in Table 3 (Materials and Methods), and agarose was added to induce self-gelation, resulting in a brown-colored hydrogel (Figure S3).

Figure 2. (a) Schematic illustration of HG-SA synthesis. (b) SEM images of hydrogels before (left) and after SA loading (right).

Figure 2. (a) Schematic illustration of HG-SA synthesis. (b) SEM images of hydrogels before (left) and after SA loading (right).

The successful incorporation of SA into the porous lignin–chitosan hydrogel was further confirmed by SEM analysis (Figure 3b). Following SA incorporation, the porous cavities observed in the pristine hydrogel were no longer visible, indicating that the pores were occupied by the incorporated SA. The incorporation of SA into the hydrogel matrix likely involves three main interactions. First, electrostatic interactions occur between the deprotonated carboxyl group of SA and the protonated amine groups of chitosan [38]. Second, hydrogen bonding and van der Waals interactions contribute to the affinity between SA and the hydrogel matrix. Third, π–π stacking interactions occur between the aromatic ring of SA and the aromatic moieties of lignin [43].

2.3. Mechanical Properties of Hydrogels

The viscoelasticity of HG-B2 and HG-SA1 was assessed using oscillatory rheology (Figure 3). Both hydrogels exhibited a well-defined linear viscoelastic region at low strain amplitudes (approximately 0.1 – 1%). This behavior indicates that applied stress did not compromise the structural stability of the polymeric network. The storage modulus (G’) of HG-B2 was consistently higher than that of HG-SA1, suggesting a stronger, more rigid network in the blank hydrogel. As strain increased to 10%, both hydrogels showed a decrease in G’, indicating progressive disruption of the hydrogel network. HG-B2 exhibited a more abrupt drop in modulus, whereas HG-SA1 showed a more gradual decline, implying that SA incorporation reduced network stiffness but improved deformation tolerance. The hydroxyl and carboxylate groups of SA often disrupt hydrogen bonding and electrostatic interactions between lignin and chitosan. This competitive binding reduces intermolecular crosslinks, leading to a looser network and ultimately a softer hydrogel [44]. These hydrogel characteristics enhance polymer network hydration, increasing chain mobility and facilitating the diffusion of the encapsulated drug [25].

2.4. In Vitro SA Release Kinetics Profile

The release behavior of various HG-SA with different composition ratios between the hydrogel matrix and SA was investigated under simulated skin conditions (PBS, pH 5.5, 32 °C). The cumulative release was calculated using Eq. 2, and the calibration curve for determining drug concentration is presented in Figure S4. As depicted in Figure 4a, HG-SA3 and HG-SA4 exhibited poor cumulative release, reaching only 15-17 % after 6 h, while HG-SA2 achieved a moderate cumulative release of 49%. A higher SA loading can occupy the internal pores of the hydrogel, thereby reducing the free volume available for water penetration [45]. In addition, the hydrophobic nature of SA may interfere with hydrogen bonding between water and the polymer network, decreasing the affinity for water and consequently reducing its swelling ratio [46]. The decrease in swelling entraps the drug within the hydrogel pores, resulting in a lower cumulative release rate [43]. In contrast, HG-SA1 exhibited a remarkable drug release rate of 96%. A lower drug loading combined with a higher hydrogel matrix content promotes molecular dispersion and complete encapsulation of the drug within the polymer network. This minimizes drug-drug interactions, allowing for higher drug diffusion and release fraction [47]. The release profile showed an initial moderate burst within the first hour, likely due to surface-associated drug molecules, followed by a sustained, controlled release phase extending up to 6 hours.

To better understand the drug-release mechanism from the lignin-chitosan hydrogel, the release kinetics were evaluated using two different mathematical models: Higuchi and Korsmeyer-Peppas. After fitting the release data for various HG-SAs to the models using Eq. 3 and 4 (Figure 4b and Figure 4c), the R² for each HG-SA was obtained. From the linear curve, both models exhibited good fitting with R² values >0.95. Nevertheless, the Korsmeyer-Peppas model was the best fit by comparing the R² (Table 2).

The Higuchi model confirms that the drug release is diffusion-controlled from the open network of the hydrogel pores [48,49]. Given that the Korsmeyer-Peppas model is well-fitted to experimental data with higher R² values, the kinetic mechanism was further observed from the n value that was obtained from the slope. As shown in Table 2, all HG-SA exhibited release exponent (n) values in the range of 0.45 < n < 0.89, suggesting a non-Fickian diffusion mechanism. This anomalous transport mechanism indicates that drug release is controlled by the combined effects of drug diffusion through the hydrogel matrix and polymer network relaxation, as illustrated in Figure 4d [50]. These findings demonstrate that diffusion is the dominant mechanism underlying drug release, as supported by both kinetic models. This short-term sustained release behavior is ideal for anti-acne patch applications, ensuring prolonged drug availability at the site of action, reducing dosing frequency, and minimizing potential skin irritation associated with rapid drug release.

2.5. Antibacterial Activity Against C. acnes

Acne symptoms are associated with abnormal proliferation of C. acnes bacteria in hair follicles and inflammation [51]. Therefore, it is necessary to investigate the antibacterial activity of the hydrogel against acne-causing bacteria. To evaluate antimicrobial activity against C. acnes, the disk diffusion method was used with ciprofloxacin as the positive control. From the antibacterial ring shown in Figure 5a, HG-B2 and HG-SA1 exhibited significantly greater inhibitory activity than lignin and chitosan alone.

The measured inhibition zone diameter of the treated HG-B2 and HG-SA1 was 10.5 mm and 12.7 mm, respectively (Figure 5b). HG-B2 itself exhibited remarkable antibacterial activity due to the antimicrobial properties of lignin and chitosan, owing to the hydroxyl groups that can inactivate bacteria by damaging cell walls [52,53]. Moreover, lignin proved to have ROS scavenging ability that could suppress the activation of immature bacterial cells [54]. Thus, incorporating the anti-inflammatory SA into the lignin-chitosan hydrogel synergistically enhances antibacterial activity, resulting in a higher inhibition zone of HG-SA1. In contrast, lignin and chitosan alone showed lower inhibition zone diameters of 3.6 mm and 4.1 mm, respectively. These findings suggest that HG-SA1 functions as a controlled drug delivery system that amplifies antibacterial efficacy beyond that of lignin and chitosan alone.

2.6. Molecular Docking and Dynamics Simulations

The disk-diffusion assay established that the SA-containing hydrogel HG-SA1 produced a larger inhibition zone against C. acnes than the corresponding SA-free hydrogel HG-B2 (12.7 vs 10.5 mm). This experimental result indicates that incorporation of SA contributes to the antibacterial response, but it does not identify the molecular process responsible. The computational analysis was therefore used to examine whether SA can associate with 6KHM-A, the open form of C. acnes lipase, at sites and with contacts that could plausibly influence ligand access or the catalytic environment. Targeted docking first tested recovery of the five predefined SA-binding regions. Molecular dynamics (MD) simulations over 100 ns were then used to assess whether complexes initiated from these poses retained comparable protein compactness, solvent exposure, backbone deviation, and local flexibility. Together, these analyses provide a mechanistic rationale that can be interpreted alongside the wet-lab result; they do not replace a direct lipase-inhibition CB-Dock2 predicts ligand-binding cavities using the CurPocket algorithm, which detects concave regions on the protein surface based on local surface curvature.

Based on these simulation results, the five best-binding regions are identified for subsequent targeted docking. Targeted docking was used to test whether SA could reproducibly occupy the five predefined binding regions on 6KHM-A under a common QuickVina-W protocol. Across the triplicate runs, the recovered poses were like, or overlapped with, the five reference binding modes (Figure S5a-e), where each binding mode denotes a distinct SA location and orientation on the protein. The mean docking affinities for modes 1-5 were -5.7, -4.3 ± 0.0578, -4.1 ± 0.0578, -4.0, and -3.6 kcal/mol, respectively. Thus, SA was not restricted to a single pose, although mode 1 had the most favorable score within this protocol and is illustrated in Figure 6a. Recovery across the predefined regions supports the structural feasibility of SA association with 6KHM-A. Docking scores are comparative hypotheses within this calculation, however, and do not by themselves demonstrate enzyme inhibition. The subsequent interaction analysis therefore focused on binding mode 1 to examine the predicted contacts between SA and 6KHM-A.

The docking poses were next examined in three and two dimensions to determine whether the predicted association places SA near residues relevant to lipase function. The interaction map for the complex shows hydrogen-bond contacts involving SA and Ser114, one of the catalytic residues of 6KHM-A, together with contacts to residues in or near the lid domain (Figure 6b). A pi-pi T-shaped interaction with Trp192 further contributes to the predicted noncovalent contact network. Across modes 1-4, SA also contacted lid-domain residues (Arg144-Asn231), a region that regulates access to the lipase active site. These contacts provide a plausible structural basis by which SA binding could perturb substrate access or the local catalytic environment. They should be interpreted as a mechanism hypothesis consistent with the antibacterial effect of HG-SA1, rather than as direct evidence of competitive inhibition.

The docking results identify candidate SA-6KHM-A complexes, but a favorable pose is only useful if the protein does not undergo a large nonspecific structural disruption after simulation begins. MD simulation (100 ns) was therefore performed for the complex. Radius of gyration (Rg) was examined first because it reports protein compactness over time [55]. Across the simulations, the 6KHM-A systems showed relatively stable Rg values of 1.84-1.85 nm, and the selected-complex trajectory is shown in Figure 6c. These small changes indicate that SA binding in the tested modes was not associated with large global expansion or collapse of 6KHM-A during the simulated period. The result supports interpretation of the docking contacts within an intact protein framework, although Rg does not measure ligand residence or inhibition.

Given that compactness alone cannot indicate whether a ligand-bound complex undergoes substantial changes in solvent exposure, solvent-accessible surface area (SASA) was assessed over the same trajectories. The SA-6KHM-A systems showed stable SASA profiles, with average values of 162-167 nm² across 100 ns (Figure 6d). This limited variation complemented the Rg result and suggests that SA binding did not produce a major global rearrangement of the protein surface. In the context of the wet-lab experiment, the stable Rg and SASA profiles strengthen the view that the docking contacts are compatible with a structurally maintained lipase conformation, but they do not identify the extent of enzyme inhibition.

Backbone root-mean-square deviation (RMSD) was then used to determine whether the complexes remained close to their initial conformations after equilibration. All SA-6KHM-A systems showed relatively stable C-alpha RMSD trajectories over 100 ns, with average values of 0.14-0.19 nm (Figure 6e). Together with Rg and SASA, these values indicate that the protein retained its overall conformation while accommodating SA in the tested binding modes. This behavior is consistent with a stable protein-ligand association that could support the observed antibacterial effect of the SA-containing hydrogel, while direct ligand-RMSD, contact-occupancy, and enzyme-activity data are still required to demonstrate a persistent inhibitory complex.

Global stability metrics do not reveal whether SA association changes flexibility near the lipase lid and the alpha7-alpha8 region that controls access to the active site. RMSF analysis of backbone C-alpha atoms was therefore used to compare local fluctuations (Figure 6f). Most residues displayed low to moderate fluctuations, with an overall mean RMSF of 0.0816 nm, supporting the structural stability of the SA-bound complex. The lid domain showed a mean RMSF of 0.0976 nm, whereas the alpha7 helix and alpha7-alpha8 loop displayed higher local flexibility, with mean RMSF values of 0.1300 and 0.1117 nm, respectively. Prominent fluctuations at residues 204 and 212, reaching 0.2131 and 0.2410 nm, highlight the dynamic character of the alpha7-alpha8 region, which is associated with the lid architecture and access to the lipase active site. Phe176 showed a moderate RMSF value of 0.0889 nm, while Phe211 was more flexible at 0.1788 nm, indicating that local motion in the binding mode 1 complex is concentrated more strongly around Phe211. Overall, these results show that SA binding mode 1 is accommodated within a globally stable 6KHM-A structure while preserving local dynamics in the lid-associated region. This residue-level behavior complements the docking contacts and supports a molecular basis for SA association with a lipase conformation relevant to substrate access and potential enzymatic regulation.

The combined experimental and computational evidence supports the antibacterial role of SA. The larger inhibition zone of HG-SA1 relative to HG-B2 suggests that SA enhances antibacterial activity, while molecular docking consistently predicts SA binding within 6KHM-A, including interactions near the catalytic site and the lid-domain region. The Rg, SASA, and RMSD results show that these modeled complexes can be accommodated without major global protein disruption, and the RMSF analysis identifies local regions that may respond differently to the binding pose. The data therefore support a molecular rationale in which SA released from the hydrogel may contribute to antibacterial activity through association with C. acnes lipase.

3. Conclusions

This study successfully developed a lignin:chitosan:agarose composite hydrogel as a short-term sustained-release platform for salicylic acid (SA) delivery in topical acne treatment. The optimized hydrogel (HG-B2) exhibited a stable three-dimensional porous network, high swelling capacity, and suitable viscoelastic properties, making it the optimal hydrogel matrix for SA incorporation. The SA-loaded hydrogel (HG-SA1) demonstrated efficient in situ drug loading, an initial burst release followed by sustained release over 6 h through a diffusion mechanism. It also demonstrated enhanced antibacterial activity against C. acnes compared with the blank hydrogel, confirming the contribution of SA to the antimicrobial performance. Molecular docking and molecular dynamics simulations complemented these experimental findings by demonstrating that SA can reproducibly associate with the catalytic and lid-domain regions of C. acnes lipase while maintaining a stable protein–ligand complex throughout the simulation. These computational results provide a plausible molecular basis for the enhanced antibacterial activity observed with HG-SA1, suggesting that SA released from the hydrogel may contribute to the regulation of bacterial lipases. Hence, the integration of experimental characterization with molecular simulations demonstrates that the lignin:chitosan:agarose hydrogel is a promising natural-polymer platform for controlled SA delivery and provides mechanistic insight into its potential antibacterial action, supporting its further development as an anti-acne patch.

4. Materials and Methods

4.1. Materials

Kraft lignin powder was purchased from Sigma-Aldrich. Chitosan (medium molecular weight, degree of deacetylation approximately 85%) was obtained from Himedia. Agarose (molecular biology grade) and analytical-grade salicylic acid (SA), acetic acid, Na₂HPO₄, KH₂PO₄, KCl, NaCl, HCl, and NaOH were sourced from Merck. All chemicals and solvents were used as received without further purification. The bacterial strain used in this study was obtained from the Bacteriology Laboratory culture collection of the National Research and Innovation Agency (BRIN), namely C. acnes ATCC 11827. Deionized water was used throughout all experiments: nutrient Broth (NB) medium, Mueller-Hinton Agar (MHA) medium, and ciprofloxacin antibiotic.

4.2. Lignin-Chitosan Hydrogels (HG-B) Formula Optimization

Hydrogel formulation optimization was carried out by preparing hydrogels with three different lignin-to-chitosan composition ratios, namely HG-B1 (9:1), HG-B2 (1:1), and HG-B3 (1:9). In this process, 0.4 % (w/v) chitosan was dissolved in 2% (v/v) acetic acid. Separately, 0.4 g of lignin powder was dispersed in 100 mL of distilled water and sonicated for 15 minutes to ensure a uniform suspension. Agarose was dissolved in distilled water by heating to approximately 90 °C to prepare a 4% (w/v) solution. The lignin dispersion was then slowly added to the chitosan solution under continuous stirring according to the predetermined lignin-to-chitosan ratios. Then, the agarose solution was added to the mixture, heated to 60°C, and stirred for 20 min. The detailed compositions of lignin, chitosan, and agarose used in each formulation are presented in Table S1. Gelation time of the three hydrogels was also studied at 5°C in a precooled water bath.

4.3. SA-Loaded Lignin-Chitosan Hydrogels (HG-SA) Preparation

The hydrogel formulation with the optimal lignin-to-chitosan ratio (HG-B2), determined in the previous optimization experiment, was used to prepare the SA-loaded hydrogel. Initially, SA was added to 1 mL of the total lignin-chitosan mixture. Then, agarose was added slowly and heated at 60°C with continuous stirring for 20 min. The final mixture was allowed to cool to room temperature to induce gelation, and the resulting hydrogel was stored at 4 °C until further analysis. This whole process resulted in the synthesis of various HG-SA, as shown in Table 3.

4.4. Characterization

Fourier-transform infrared (FTIR) spectroscopy was employed to characterize the chemical structure and interactions within the hydrogel matrix. Dried hydrogel samples were analyzed using a spectrometer equipped with an attenuated total reflectance (ATR) accessory, scanning over the range of 4000 to 400 cm^-1 with a resolution of 4 cm^-1 and 32 scans per sample. To observe the microstructure and porosity, scanning electron microscopy (SEM) was conducted on freeze-dried hydrogel samples that were sputter-coated with gold. Images captured at various magnifications revealed the surface morphology and pore distribution. Complementary to SEM, Brunauer–Emmett–Teller (BET) surface area analysis was performed using nitrogen adsorption-desorption isotherms at 77 K to quantify the specific surface area and pore volume of the hydrogel. Samples were degassed under vacuum at 60 °C for 12 hours before measurement, and the Barrett-Joyner-Halenda (BJH) method was applied to determine pore size distribution. The rheological measurement of the hydrogel was performed on a Discovery core rheometer equipped with a 20 mm parallel plate (1000 µm set gap), a peltier plate, stainless steel, and a solvent trap. The oscillatory strain measurement was performed over a strain range of 0.01-100% at a 10.0 rad/s angular frequency at 25°C. The Genesys 150 UV-Vis spectrophotometer from Thermo Scientific was used to measure SA absorbance in the drug release study.

4.5. Hydrogel Swelling Test

The swelling test of various HG-B (Table S1) was performed by immersing the freeze-dried hydrogel in 1 mL of distilled water, and the weight was measured over 48 h. The swelling ratio was determined using Eq. (1) with the following formula: [56]

$$\mathrm{s}\mathrm{w}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}(Q\%)={\displaystyle \frac{W_s-W_d}{W_d}}\times 100$$

(1)

Where W_s and W_d are the swollen and dried hydrogel, respectively.

4.6. In vitro SA Release Study

To determine the release profile of SA from the hydrogel, in vitro drug-release studies were conducted by placing various HG-SA formulations (Table 3) into a 50 mL tube. The hydrogels were immersed in phosphate-buffered saline (PBS, pH 5.5) maintained at 32 ± 0.5 °C to simulate skin surface temperature and gently agitated at 100 rpm. Aliquots of the release medium were withdrawn at predetermined time intervals up to 6 hours. The concentration of SA in each aliquot was quantified by UV-Vis spectrophotometry at 297 nm, and cumulative release percentages were calculated and plotted to evaluate release kinetics according to the following formula: [38]

$$Cumulativerelease(\%)={\displaystyle \frac{Q_t}{Q_m}}\times 100$$

(2)

Where Qt represents the amount of SA released from the hydrogels at time t, and Q_m denotes the total amount of SA loaded onto the hydrogel.

The release kinetics of various HG-SA were determined by the following models:

• Higuchi models [48]

$$Qt=KH\surd t$$

(3)

Where the K_H represents the Higuchi dissolution constant, and t represents time.

• Korsmeyer-Peppas [57]

$${\displaystyle \frac{M_t}{M_{\infty }}}=K{t}^n$$

(4)

Where M_t/M_∞ represents the fraction of drug released at time t₁, K and n represent the rate constant and the release exponent, respectively.

4.7. Antibacterial Activity Assay

The materials required include four samples (lignin solution, chitosan solution, blank hydrogel (HG-B2), and HG-SA1, NB medium, MHA medium, ciprofloxacin antibiotic, blank antibiotic discs, and C. acnes bacterial isolate. Antibacterial activity in this study was tested using the agar disc diffusion method. This method begins with preparing MHA medium for antibacterial testing. A total of 200 µl of microbes in NB medium was spread on the surface of solid media (MHA) using a sterile spreader. Antimicrobial testing was carried out using discs soaked in lignin and chitosan solutions for 30 min each. In addition to these two materials, samples of HG-B2 and HG-SA1 were also used. Control solutions in this test included positive and negative controls. The positive control was the antibiotic ciprofloxacin, and the negative control was sterile distilled water. After that, incubation was carried out at 37°C for 18-24 h. After incubation, an inhibition zone formed, and its diameter was measured with a caliper. The test was performed twice to determine the size of the inhibition zone formed. The formed inhibition zone indicates antibacterial activity against C. acnes.

4.8. Preparation of Salicylic Acid and 6KHM-A Lipase Structures

SA structures and reference coordinates were extracted from pre-docked SA binding modes on chain A of the 6KHM macromolecule (6KHM-A) [58]. 6KHM-A is the open form of a C. acnes lipase, with the catalytic residues Ser114, Asp252, and His285 accessible to external ligands in the open conformation. Chain A was selected for all docking and molecular dynamics analyses because 6KHM is a dimeric protein and chain A is structurally equivalent to chain B [59]. AutoDockTools 1.5.7 from MGLTools was used to prepare SA and 6KHM-A in PDBQT format for targeted docking [60]. Ligand preparation included Gasteiger charge assignment [61] and removal of nonpolar hydrogens.

4.9. Molecular Docking Study

The preliminary step of the molecular docking procedure involves cavity detection, which is performed using the web-based application CB-Dock2 [62]. CB-Dock2 searches for concave surfaces to detect cavities. Based on simulation results, the five best cavities were identified as recommended for protein-ligand binding regions [63]. At the next step, targeted docking of the prepared SA ligand to the 6KHM-A macromolecule was carried out using QuickVina-W [64] in triplicate, with an exhaustiveness value of 32 and grid boxes of 14 x 14 x 14 A3. The five docking grids were centered on the reference SA binding modes so that each run sampled one predefined binding region. This setup was used to compare whether SA could reproduce the reference poses and to identify the most consistent poses for subsequent molecular dynamics analysis. Docking poses were visualized with UCSF ChimeraX 1.7.1 [65], and protein-ligand interactions were inspected with BIOVIA Discovery Studio Visualizer 2021.

4.10. Molecular Dynamics Simulations

Reference SA structures and coordinates were extracted as Sybyl MOL2 (.mol2) files. Ligand topology files were generated using AnteChamber PYthon Parser interfacE (ACPYPE) [66] for compatibility with the general AMBER force field 2 (GAFF2) [67]. The 6KHM-A topology was generated in GROMACS 2021.2 [68] using the AMBER99SB-ILDN all-atom force field [69] and the TIP3P water model.

MD simulations were performed in triplicate for 100 ns for each SA-6KHM-A complex representing binding modes 1-5 using GROMACS 2021.2. Each system was placed in a dodecahedron box maintained at 310 K and 1 bar, containing 11,775; 11,778; 11,775; 11,778; and 11,783 water molecules for binding modes 1-5, respectively, and neutralized with 19 Na⁺ ions. The simulation timestep was set to 0.002 ps. Before production, each system was subjected to energy minimization, 100 ps Number of particles, volume, and temperature (NVT) equilibration, and 100 ps Number of particles, pressure, and temperature (NPT) equilibration. Long-range electrostatics were treated using the particle mesh Ewald (PME) method [70], hydrogen bonds were constrained with the LINCS algorithm [71], temperature was coupled with the modified Berendsen thermostat [72], and pressure was coupled with the Parrinello-Rahman method [73]. Radius of gyration (Rg), solvent-accessible surface area (SASA), root-mean-square deviation (RMSD), and root-mean-square fluctuation (RMSF) were calculated using gmx gyrate, gmx sasa, gmx rms, and gmx rmsf, respectively.