2.1. Study Design and Specimen Collection

This cross-sectional study included 660 sputum and throat swab samples collected from primary care facilities and tertiary hospitals in Pakistan. It ensured that the patients had not undergone any previous antibiotic treatment. The study received ethical approval from the institutional review committee of the University of Lahore, Pakistan, and adhered to the World Medical Association’s Declaration of Helsinki guidelines [26]. Sterile containers with Amies medium and swabs were employed to transport the samples to the laboratory. Upon arrival at the laboratory, the samples were inoculated onto nutrient agar and incubated overnight at 37°C. Well-isolated single colonies were selected and transferred to agar plates (blood, chocolate, and MacConkey), which were then incubated at 37°C for 24–48 h. The suspected colonies were subsequently subcultured for additional examination and identification.

2.2. Identification of Bacterial Isolates

P. aeruginosa isolates were identified by colony morphology and standard biochemical tests. API 20NE (bioMérieux, Craponne, France) was employed to identify P. aeruginosa isolates [27]. In addition, the isolates underwent molecular characterization to verify the presence of P. aeruginosa. The conventional boiling method was employed to extract DNA from the bacterial samples, which was then subjected to PCR amplification. The OprI gene (249 bp) was amplified using the primers OprI-F (5 ^′-ATGAACAACGTTCTGAAATTCTCTGCT-3 ^′) and OprI-R (5 ^′-CTTGCGGCTGGCTTTTTCCAG-3 ^′), while the OprL gene (504 bp) was amplified with the primers OprL-F (5 ^′-ATGGAAATGCTGAAATTCGGC-3 ^′) and OprL-R (5 ^′-CTTCTTCAGCTCGACGCGACG-3 ^′) [28–30]. The PCR reactions were performed at a total volume of 25 μL. This mixture consisted of 11 μL of DNase, 8 μL of 2X PCR Master Mix with 1.5 mM MgCl₂, 0.5 μL of each primer, and 5 μL of DNA template. The thermal cycling procedure started with a 5-min initial denaturation step at 94°C. This was followed by 30 repetitions of a three-step cycle: denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and extension at 72°C for 1 min. The procedure concluded with a final extension phase lasting 10 min at 72°C. Controls included positive samples (P. aeruginosa DNA) and negative samples (water). The PCR products were separated using 1% agarose gels in TBE buffer (40 mM Tris, 20 mM boric acid, and 1 mM EDTA at pH 8.3) and visualized under UV light [31].

2.3. Antimicrobial Susceptibility Testing

Antimicrobial susceptibility against all the major classes of antibiotics was assessed using the Kirby–Bauer disc diffusion method. A variety of standard antibiotic discs (Oxoid, United Kingdom) were used, including imipenem (10 μg), meropenem (10 μg), piperacillin/tazobactam (100/10 μg), ticarcillin/clavulanic acid (75/10 μg), cefepime (30 μg), cefoperazone (30 μg), cefotaxime (30 μg), ceftazidime (30 μg), ceftriaxone (5 μg), amikacin (30 μg), gentamicin (10 μg), tobramycin (10 μg), tigecycline (4 μg), ciprofloxacin (5 μg), and levofloxacin (5 μg). The ATCC 27853 strain (GenBank accession CP015117) of P. aeruginosa was employed for quality control (QC) purposes. The findings were analyzed according to the guidelines established by the Clinical and Laboratory Standards Institute (CLSI) [32]. The isolates that showed resistance against at least three antibiotic classes were declared MDR, and those against at least five classes were declared extensively drug-resistant (XDR). The isolates that resisted all the tested classes of antibiotics were considered pan-drug-resistant.

2.4. Determining the Antibacterial Efficacy of Aloe vera–Derived Phytochemicals Using Agar Well Diffusion Assay

Based on an extensive literature review, the antibacterial activity of A. vera phytochemicals, including aloin, aloe-emodin, acemannan, saponins, lupeol, salicylic acid, and cinnamic acid, was evaluated against P. aeruginosa using the agar well diffusion method. P. aeruginosa was cultured in nutrient broth at 37°C, and the bacterial culture was used when it reached the log phase of growth. Wells 6 mm in diameter were created using a sterile borer in Mueller–Hinton agar plates. A 100-μL amount of bacterial suspension was uniformly spread using a sterile cotton swab in a circular motion across the agar surface to form a bacterial lawn. Phytochemicals, such as aloin, aloe-emodin, acemannan, saponins, lupeol, salicylic acid, and cinnamic acid, were dissolved in DMSO at a concentration of 2 μg/mL, and 100 μL of each phytochemical solution was placed into the designated wells. The plates were incubated at 37°C for 24 h. All experiments were conducted in triplicate, and the inhibition zones were measured according to CLSI standards [32].

2.5. Evaluation of the Minimum Inhibitory Concentration (MIC) of Aloe vera Phytochemicals

The antibacterial activity of A. vera phytochemicals against MDR P. aeruginosa was evaluated using a MIC assay in a 96-well microtiter plate. Phytochemicals such as aloin, aloe-emodin, acemannan, saponins, lupeol, salicylic acid, and cinnamic acid were dissolved in DMSO with a 1000 μg/mL stock. Two-fold serial dilutions, ranging from 1000 to 31.25 μg/mL, were made. The MDR P. aeruginosa was grown in Mueller–Hinton broth (MHB) at 37°C until it reached the mid-logarithmic growth phase. The bacterial suspension was standardized to a 0.5 McFarland standard (1.5 × 10⁸ CFU/mL). In the microtiter plate, 100 μL of MHB was added to each well, followed by 100 μL of each phytochemical, beginning with the highest concentration and proceeding through two-fold serial dilutions. Then, 50 μL of the bacterial suspension was added to the wells containing the phytochemicals. The positive control wells contained only the bacterial suspension without any phytochemicals, while the negative controls had only MHB. The plate was incubated for 24 h at 37°C. After incubation, 20 μL of resazurin dye was introduced into each well and allowed to react for 2–4 h. A blue color indicated the absence of bacterial growth, whereas a pink color suggested active growth. The MIC was determined as the lowest phytochemical concentration that prevented visible bacterial growth, indicated by the absence of a color change. All experiments were conducted in triplicate.

2.6. Statistical Analysis

Statistical analyses were performed using SPSS version 25.0. Frequencies were determined for qualitative variables, while the mean ± standard deviation (SD) was calculated for quantitative variables. The chi-square test was employed to evaluate associations between categorical variables, and one-way analysis of variance (ANOVA) was conducted to compare the mean inhibition zones between the different phytochemicals and tigecycline. Both have a significant level determined with a p value ≤ 0.05. A heat map illustrating antibiotic resistance profiles of P. aeruginosa isolates was generated in R (4.4.2) using the heat map package.

2.7. Evaluation of Drug Profile Through ADME and Toxicity Analyses

Pharmacokinetic and absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties were assessed using the pkCSM web server (http://structure.bioc.cam.ac.uk/pkcsm) and SwissADME server (http://www.swissadme.ch; accessed on September 4, 2024). A molecular fingerprinting approach was applied to evaluate drug-likeness, based on Lipinski’s rule of five (LRF). Toxicological assessments were conducted using the ProTox-3.0 (ProTox-3.0—Prediction of Toxicity of chemicals (charite.de)) servers and StopTox (https://stoptox.mml.unc.edu/), which evaluated various toxicity endpoints using QSAR models and machine learning methods [33].

2.8. Retrieval of Ligands and Receptors

The crystal structure of wild-type AmpC protein (PDB ID: 7FF0) was obtained from the Protein Data Bank (RCSB PDB; https://www.rcsb.org; accessed on September 4, 2024) in PDB format. The structural representations of these therapeutically significant compounds were obtained from the PubChem database. ChemDraw Professional (Version 21.0) was employed to construct 2D and 3D structures of these compounds, while Discovery Studio Visualizer (v21.1.0.20298) was used for their visualization [34, 35].

2.9. Protein Preparation

The protein structure was imported from the PDB, and any structural defects in the residues were corrected. The Schrödinger suite (Maestro Wizard version 21.5) was used to process the proteins. Initially, preprocessing was performed by filling the gap between loops and side chains by applying prime job, adding hydrogen atoms, and creating disulfide bonds at a constant pH of 7.0 ± 2. Preprocessing was followed by H-bond optimization using PROPKA at an optimized pH of 7.0; subsequently, water molecules in a radius of 3.0 Å were removed. Finally, the structure was minimized using the OPLS3e force field. The scaling factor for the van der Waals (vdW) radius was adjusted to 1.0, and grid files outlining the receptor binding sites were created [36].

2.10. Ligand Preparation

The reference compound and ligands were generated using the LigPrep module of Schrödinger. Low-energy three-dimensional conformations were created with suitable chiralities and refined using the OPLS3e force field at a physiological pH of 7.2 ± 0.2. A detailed protocol by Rauf et al. was followed [37].

2.11. Protein–Ligand Docking

Protein–ligand interactions were analyzed using Maestro because of its enhanced precision. The binding site’s orientation and dimensions for ligand docking were determined by constructing a receptor grid. The Schrödinger suite was used to generate the scoring grid based on the AmpC crystal structure (Maestro Wizard version 21.5). Molecular docking was performed using Glide in Schrödinger Maestro 21.5 and employed both standard precision (SP) and extra precision (XP) approaches, along with flexible ligand sampling. The cutoff for partial charges on ligand atoms was set to 0.15, while the scaling factor for vdW radii was set to 0.80. Additionally, root-mean-square deviation (RMSD) value was also computed to validate the docking accuracy [36].

2.12. Visualization of Protein–Ligand Complexes

AmpC protein structures were visualized using UniProt and AlphaFold. Discovery Studio Visualizer 20.1 was used to visualize protein–ligand complexes and identify interactions. Docked structures were also analyzed using PyMOL (Version 2.4.0).

2.13. Density Functional Theory (DFT) Analysis

DFT computations were used to investigate the physicochemical characteristics of the chosen compounds. The RB3LYP method was employed for both optimization and frequency analysis using the Gaussian 09 W program. The energy levels of the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO), as well as the band gap, were determined using the same theoretical framework to derive additional quantum chemical parameters. The energy band gap (ΔEGap = ELUMO − EHOMO) was obtained by calculating the difference between the LUMO and HOMO energy values and visualized on Gauss View 5.0.8 software. Furthermore, the DFT method was used to evaluate the molecular electrostatic potential of the compounds by using a checkpoint (.chk) file [38]. A detailed protocol by Rauf et al. was followed [39].

3.1. Specimens and Isolates

This study included 660 patients from diverse ethnic backgrounds. Of these, 387 (58.6%) were male, and 273 (41.4%) were female. The incidence of P. aeruginosa infection was significantly higher in males (54%) compared to females (46%) (p = 0.05). The participants’ ages ranged from 1 to 95 years, with a mean age of 50.31 ± 21.113 years. The prevalence of P. aeruginosa varied by age group, with the highest prevalence in Age Group IV (61–80 years) at 31.41% (n = 89), followed by Age Groups II (21–40 years) and III (41–60 years) at 26.9% (n = 76) and 27.5% (n = 77), respectively (Table 1). The lowest prevalence was observed in Age Group I (1–20 years) at 8.3% (n = 25) and Age Group V (81–100 years) at 5.65% (n = 16). The differences in prevalence between age groups were statistically significant (p = 0.02). Regional distribution showed 18 cases (2.8%) from Azad Jammu and Kashmir, 37 cases (5.6%) from Khyber Pakhtunkhwa, 573 cases (86.8%) from Punjab, and 32 cases (4.8%) from Sindh. The highest infection rate was in the Pakhtoon ethnic group at 30% (n = 238), followed by the Sindhi (28%), Punjabi (22%), and Kashmiri (20%) groups.

3.2. Microbial Spectrum Associated With Respiratory Origins

The microbial spectrum of respiratory infections revealed the presence of 24 microbial taxa. P. aeruginosa was found to be the most prevalent, followed by Klebsiella pneumoniae and Acinetobacter baumannii. The least prevalent microbial species included Acinetobacter lwoffii, Aspergillus flavus, Aspergillus fumigatus, Citrobacter koseri, Citrobacter spp., and Enterobacter aerogenes (Table S1).

3.3. Antibiotic Susceptibility Profile of P. aeruginosa

The patterns of resistance of P. aeruginosa to all the major antibiotic classes were evaluated. All the isolates were found to be sensitive to ticarcillin/clavulanic acid, cefoperazone, cefotaxime, ceftriaxone, and tigecycline. Ciprofloxacin proved to be the most effective antibiotic, followed by cefepime, ceftazidime, gentamicin, tobramycin, imipenem, meropenem, and piperacillin/tazobactam. The antibiotic resistance patterns of the bacterial isolates were observed to be influenced by the gender, age, and provincial distribution of the patients. In general, the isolates recovered from male patients, from patients in the 41–60-year age group, and from the Punjabi ethnic group were more resistant (Table 2).

3.4. Prevalence of Coresistant P. aeruginosa

Of the 283 P. aeruginosa isolates, 177 (63.2%) were MDR and 8 (2.9%) were XDR. The distribution of coresistant isolates varied according to ethnic group, age, and gender of the patients. The highest MDR prevalence was observed in the Punjabi group, followed by the Pakhtoons, Sindhi, and Kashmiris. The prevalence of XDR isolates was highest in the Punjabi group (2.5%) and lowest in the Kashmiri and Sindhi groups (0.0%). Non-MDR isolates were most prevalent in the Punjabi group and least prevalent in the Kashmiri group. However, these differences in prevalence were not statistically significant. Moreover, male patients were observed to harbor more non-MDR and MDR isolates, but an inverse trend was observed for XDR. Furthermore, the prevalence of non-MDR and MDR isolates was observed to increase with increasing age of the patients up to 80 years. However, XDR isolates were more prevalent in the 41–60-year age group (Table 2). In addition, a cluster analysis was performed to assess both the efficacy of each antibiotic in combating P. aeruginosa and the overall sensitivity pattern across all tested antibiotics. The analysis of bacterial isolate clustering based on their AMR patterns revealed considerable diversity. The dendrogram in Figure 1 visualizes the grouping of P. aeruginosa isolates according to their AMR profile. The extent of branching in the dendrogram reflects the variation in resistance, with more diverse branches indicating greater variability between the isolates.

3.5. Antibacterial Efficacy of Aloe vera Phytochemicals Determined Using Agar Well Diffusion Assay

The antibacterial activity of A. vera phytoconstituents against the PDR strain of P. aeruginosa was determined at a concentration of 2 mg/mL. The maximum zone of inhibition of cinnamic acid was 21.10 ± 0.17 mm, while those of acemannan 18.93 ± 0.11 and lupeol were 16.9 ± 0.17 mm each, aloe-emodin was 19.93 ± 0.11 mm, and saponins were 18.93 ± 0.11 mm. The smallest inhibition zones were recorded for aloin and salicylic acid because they inhibited growth only by 15.93 ± 0.11 mm and 14.43 ± 0.11 mm, respectively (Table 3, Figure 2). Tigecycline (15 μg/Disc) is used as a standard antibiotic, showing an 11 ± 00 mm inhibition zone. The results of the ANOVA test (p < 0.0001) ensure that the differences between phytochemicals and tigecycline inhibition are highly significant (Table 4).

3.6. The MIC of Aloe vera Phytoconstituents

The MICs of A. vera phytochemicals against MDR P. aeruginosa were as follows. Lupeol had the lowest MIC of 125 μg/mL, thus making it the most potent of all the phytochemicals. Aloe-emodin, acemannan, salicylic acid, and cinnamic acid had moderate MICs, with all having values of 250 μg/mL. Steroidal saponins and aloin had the highest MIC of 500 μg/mL, thus indicating that they were the least effective compared to the other phytoconstituents (Figure 3). The antibiotic imipenem/cilastatin 100 μg/mL was used as a standard positive control. The p value is < 0.0001, suggesting statistical significance in the differences between groups. The three asterisks indicate a significant result. There is an R-squared of 0.9979, which means the model explains 99.79% of the variability in the dependent variable. A very high R-squared means the model fits the data pretty well and captures nearly all the variance in the data (Table 5).

3.7. In Silico Analysis

The in silico analysis demonstrated that mutations in multiple genes contributed to antibiotic resistance. It focused on the structure of AmpC, which is known for its role in cephalosporin resistance through the overproduction of chromosomal cephalosporinase. Bioinformatics tools were used to investigate potential resistance-causing variants. The investigation revealed three key hotspot mutations in AmpC: T96I, G183D, and E247K. The T96I mutation involves the substitution of threonine with isoleucine at position 96, while G183D involves a glycine-to-aspartic acid change at position 183. The E247K mutation replaces glutamic acid with lysine at position 247. The three-dimensional structure of the AmpC protein (PDB ID 7FF0) was visualized using UniProt and AlphaFold to enhance the understanding of how these mutations affected antibiotic resistance (Figure S1). Docking was performed against the wild type and mutated AmpC structures using the cocrystal ligand alongside seven additional Aloe vera–derived phytochemicals: aloin, aloe-emodin, acemannan, saponin, lupeol, salicylic acid, and cinnamic acid selected based on virtual screening (Table 6).

Table 6. Two-dimensional (2D) and three-dimensional (3D) structures and Simplified Molecular Input Line Entry System (SMILES) strings of the selected phytochemicals of Aloe vera.

Code	Phytochemical	2D structure	3D structure	SMILES
PA-1	Aloin			C1=CC2=C(C(=C1)O)C(=O)C3=C(C2C4C(C(C(C(O4)CO)O)O)O)C=C(C=C3O)CO
PA-2	Aloe-emodin			C1=CC2=C(C(=C1)O)C(=O)C3=C(C2=O)C=C(C=C3O)CO
PA-3	Acemannan			CC(=O)NC1C(C(C(OC1OC2C(OC(C(C2OC(=O)C)O)OC3C(OC(C(C3OC(=O)C)O)OC4C(OC(C(C4OC(=O)C)O)OC)CO)CO)CO)CO)OC5C(C(C(C(O5)CO)OC6C(C(C(C(O6)C(=O)[O-])OC7C(C(C(C(O7)CO)OC8C(C(C(C(O8)CO)OC)OC(=O)C)O)OC(=O)C)O)OC(=O)C)O)OC(=O)C)O)O
PA-4	Saponin			CC1(C2CCC3(C(C2(CCC1OC4C(C(C(CO4)OC5C(C(C(CO5)O)O)O)OC6C(C(C(C(O6)CO)O)O)O)OC7C(C(C(C(O7)CO)O)O)OC8C(C(C(C(O8)CO)O)O)O)C)CCC91C3(CC(C2(C9CC(CC2)(C)C=O)CO1)O)C)C)C
PA-5	Lupeol			CC(=C)C1CCC2(C1C3CCC4C5(CCC(C(C5CCC4(C3(CC2)C)C)(C)C)O)C)C
PA-6	Salicylic acid			C1=CC=C(C(=C1)C(=O)O)O
PA-7	Cinnamic acid			C1=CC=C(C=C1)C=CC(=O)O

3.8. Pharmacokinetics, Drug-Likeness, and Toxicity

The key bioactive compounds of A. vera were examined, with LRF applied to assess physicochemical properties and drug-likeness. The compounds, PA-2, PA-6, and PA-7, showed no violations of Lipinski’s criteria, indicating that they possessed favorable drug-like qualities and the potential for oral bioavailability. PA-2 (aloe-emodin) emerged as a particularly promising phytochemical due to its low molecular weight (270.24 g/mol), favorable hydrophilicity (logP of 1.96), and optimal hydrogen bonding capacity (five acceptors and three donors). PA-6 (salicylic acid) was categorized as soluble according to both the estimated solubility (ESOL) and Silicos-it solubility models, with an ESOL logS of −2.5 and a Silicos-it logP of 0.74. Aloe-emodin was found to be moderately soluble, with an ESOL logS of −3.04 and a Silicos-it logP of 2.42. The compounds exhibited diverse logKp (cm/s) values in terms of permeability, which suggested variability in their ability to penetrate membranes. PA-2 and PA-6 showed moderately high permeability, with logKp values of −6.66 and −5.54, respectively, whereas lupeol demonstrated the highest permeability with a logKp value of −1.9, which suggested its stronger potential for absorption through biological membranes compared to the other compounds.

In terms of bioavailability, salicylic acid and cinnamic acid exhibited the highest scores of 0.85, indicating excellent potential for absorption and therapeutic effectiveness. The bioavailability score of 0.55 for aloe-emodin also suggested its favorable systemic distribution and therapeutic potential. The compounds varied in synthetic accessibility, with salicylic acid and cinnamic acid being the easiest to synthesize, having scores of 1 and 1.67, respectively. This analysis identified PA-2, PA-6, and PA-7 as the most promising candidates for further research based on their solubility, permeability, drug-likeness, bioavailability, and ease of synthesis (Table 7). Drug metabolism analysis, focusing on the cytochrome P450 enzyme system, revealed that PA-2 and PA-7 inhibited CYP1A2, while the other compounds did not inhibit the enzyme. Notably, PA-1, PA-3, PA-4, PA-5, and the cocrystal ligand were identified as substrates for CYP3A4, indicating a higher probability of being metabolized by this enzyme. None of the compounds inhibited CYP2C19, CYP2C9, CYP2D6, or CYP3A4, suggesting a lower risk of significant drug–drug interactions through these pathways.

The BOILED-Egg model provides valuable information about how compounds are absorbed and distributed. The model’s white region indicates favorable passive absorption in the gut, which suggests a high likelihood of oral bioavailability. The yellow area, which resembles an egg yolk, points to a strong possibility of crossing the blood–brain barrier (BBB), which implies potential involvement in central nervous system functions. The blue color identifies compounds that are substrates for P-glycoprotein (PGP+), a transporter that facilitates drug efflux, while red denotes compounds that are not substrates (PGP−). Among the key compounds, aloe-emodin fell within the white zone, indicating strong gastrointestinal absorption. Lupeol and salicylic acid were found within the yolk region, indicating a strong likelihood of crossing the BBB. In contrast, aloin, saponin, and cinnamic acid were positioned outside the egg, suggesting reduced capacity for both gastrointestinal absorption and BBB penetration (Figure S2).

The chosen phytochemicals were evaluated for acute toxicity and irritation characteristics based on multiple criteria, including inhalation, oral, and dermal toxicity, along with skin and eye irritation, using the StopTox tool (Table S2). All phytochemicals demonstrated no toxicity in terms of acute inhalation. However, salicylic acid exhibited both acute oral and dermal toxicity. Aloe-emodin and cinnamic acid were identified as toxic with regard to eye irritation and corrosion but were otherwise nontoxic across the other parameters. Cytotoxicity and hepatotoxicity evaluations were performed using the Protox 3.0 online platform, which estimated the potential toxicity of the chosen phytochemicals. According to these assessments, none of the compounds was anticipated to pose any major toxicity issues. Cinnamic acid, in particular, showed the most favorable outcomes, exhibiting minimal hepatotoxicity, neurotoxicity, and mutagenicity (Table S3). The ADMET and toxicity radar charts of the A. vera–derived phytochemicals are shown in Table S4.

3.9. Docking Studies

A molecular docking study was conducted to evaluate the interaction between the selected bioactive compounds from A. vera and the binding site of the AmpC protein in P. aeruginosa. Maestro was employed to dock the phytochemicals alongside the cocrystal ligand against both wild-type and mutant-type AmpC structures. To verify the docking outcomes, the structure was redocked with the cocrystal ligand, yielding an RMSD of 0.2002 Å for the redocked reference compound, which suggested a close structural match between the two positions in the AmpC binding site (Figure S3). All compounds exhibited different levels of interaction with the target protein in both the wild-type and mutated forms. In terms of binding affinity, compounds with more negative docking scores typically suggest a stronger attachment to the target site. Aloe-emodin, salicylic acid, and cinnamic acid showed significant docking scores of −7.97, −7.264, and −6.498 kcal/mol, respectively (Table 8), which highlighted their favorable binding interactions. The AmpC protein contained specific amino acids that demonstrated notable binding interactions with these compounds, such as F72, L66, A89, T132, R128, T133, L129, E74, and S88. The molecular visualization tools in PyMOL were employed to present the docking results for the cocrystal ligand and aloe-emodin, as illustrated in Figure 4, which highlights the interactions between the phytochemicals and the AmpC protein.

3.10. DFT Analysis of the Selected Ligands and Cocrystal Ligand

DFT analysis was performed to determine the electronic properties of the selected phytochemicals in the solvent phase (water) using the RB3LYP calculation technique. Salicylic acid and aloe-emodin were selected for DFT analysis based on the following criteria: ADMET profiling, bioactivity evaluation, and the interactions and binding affinities of the compounds. Of the ligands evaluated, aloe-emodin was the most reactive, with the smallest energy gap (0.12765 a.u.) and the highest softness value (7.835 eV), indicating that it was less stable but more flexible in terms of reactivity. The energy gap (ΔEGap) between HOMO and LUMO indicated both chemical stability and reactivity. While salicylic acid was more stable than aloe-emodin due to its larger energy gap (0.18371 a.u.), it had the highest electron affinity, as evidenced by the lowest LUMO energy (−0.04856 a.u.), making it more likely to absorb electrons. Aloe-emodin also had intermediate values for both HOMO and LUMO (Figure 5), suggesting a balanced reactivity profile compared to the higher electron-absorbing potential of salicylic acid (Table S5).