All chemical reagents and biological materials used in this study were obtained from commercial suppliers with specified purity grades. The methicillin-resistant Staphylococcus aureus strain (ATCC 43300) was acquired from the American Type Culture Collection (ATCC, USA). The strain was cultured in Tryptic Soy Broth (TSB) liquid medium and on Tryptic Soy Agar (TSA) solid medium. Both media along with vancomycin HCl (VAN) and crystal violet, were procured from Dalian Meilun Biotechnology Co., Ltd (Dalian, China), which also supplied the dimethyl sulfoxide (DMSO). CellTiter-Glo^® 2.0 was purchased from Promega Biotechnology Co., Ltd (Beijing, China). LIVE/DEAD™ BacLight™ Bacterial Viability Assay Kit (L7007) was purchased from Thermo Fisher Scientific Co., Ltd (Shanghai, China). Glutaraldehyde 2.5% (EM grade) was purchased from Beijing Solarbio Science & Technology Co., Ltd (Beijing, China). All other organic solvents including ethanol, were of analytical grade.

MRSA susceptibility to JR20 was assessed using the filter disc diffusion and microbroth dilution methods.

The anti-biofilm efficacy of JR20 was assessed using both crystal violet staining and dual-fluorescence labeling to evaluate its capacity to inhibit and disrupt fungal biofilm formation. Biofilms play a critical role in microbial colonization and represent a key mechanism contributing to drug resistance. These experimental approaches provide insight into the compound's potential in combating biofilm-associated antifungal resistance.

Intracellular ATP levels in MRSA were quantified using the CellTiter-Glo^® 2.0 luminescent assay (Promega Biotechnology, USA). The protocol was performed in strict accordance with the manufacturer's instructions, with minor adaptations to accommodate bacterial culture conditions. In this assay, ATP derived from metabolically active cells catalyzes a luciferase-based reaction, generating a luminescent signal proportional to the ATP concentration (Zhang et al., 2023).

We specifically selected this luminescence-based method over traditional colorimetric ATP assays because the inherent turbidity of dense bacterial suspensions can significantly interfere with optical density (OD) readings, leading to potential inaccuracies in absorbance-based quantification. The bioluminescent approach effectively circumvents this optical interference, providing a more reliable and specific measurement of ATP content within bacterial samples (Liang et al., 2022).

MRSA cultures in the logarithmic growth phase were plated into a 96-well plate at a concentration of 1–5 × 10⁷ CFU/mL at a volume of 100 μL per well. Different concentration gradients of JR20 were established alongside positive drug control VAN and negative DMSO control. All groups were incubated with the bacterial suspension for 24 h before removal. The prepared CellTiter-Glo^® 2.0 working solution was then added and the samples were lysed on ice for 15 min. Luminescence was detected using a multifunctional microplate reader (Synergy H1, California, USA) with an integration time of 0.25-1 s, and data were normalized by first subtracting the background and then setting the negative control to 100%. Results are presented as Relative Luminescence Units (RLU) on the y-axis.

The crystal structure of PBP2a (PDB: 1MWT), a key resistance determinant in MRSA, was obtained from the RCSB Protein Data Bank. The co-crystallized native ligand, penicillin G, was extracted for comparative analysis. The three-dimensional structure of JR20 was built using Chem3D (PerkinElmer), subjected to hydrogenation, and its torsion angles were optimized.

The protein was prepared by removing water molecules and adding polar hydrogens. Molecular docking was performed using AutoDock Vina (version 1.5.6) to assess the binding of JR20 and the reference ligand penicillin G to PBP2a. In each simulation, 1MWT was set as the receptor. A docking grid box of appropriate dimensions was centered on the active site, and ten independent docking runs were conducted. The conformation with the most favorable (lowest) binding energy was selected for analysis. The resulting complexes were visualized and analyzed using PyMOL (Schrödinger) and Discovery Studio 2021 (Dassault Systèmes).

Statistical analysis was performed using GraphPad Prism 8.0 software (San Diego, California, USA). The experimental data were obtained from three independent replicates and visualized using line graphs and bar charts. Intergroup statistical differences were determined using the appropriate parametric or non-parametric tests.

P ≤ 0.05 was considered statistically significant. Significance levels are denoted as follows: ^* p ≤ 0.05, ^** p ≤ 0.01, ^*** p ≤ 0.001, ^****p ≤ 0.0001.

Our study revealed that the tetrahydrofuran ring in sesamin can be selectively cleaved to generate arylnaphthalene lignan analogs (Wu et al., 2024; Zhang et al., 2025). In the synthetic sequence, treatment of sesamin with AlCl₃ induces ring opening of the tetrahydrofuran moiety, affording arylnaphthalene lignan 2. Subsequent deacetylation of compound 2 gives compound 3, which is further oxidized with Dess-Martin periodinane to afford the target compound JR20 (Scheme 1).

The anti-MRSA activity of JR20 was quantitatively and qualitatively evaluated through IC₅₀ determination and disk diffusion assays. As shown in Figure 1A, JR20 exhibited an IC₅₀ value of ~20 μg/mL against MRSA. In the agar diffusion test, distinct zones of inhibition were observed, with diameters positively correlated with drug concentration. At a dose of 10 μg, JR20 produced an average inhibition zone diameter of 16.9 mm, which was comparable to that of the vancomycin positive control (14.86 μg) and falls within the clinically defined highly sensitive range (Figure 1B). These results collectively demonstrate the potent in vitro anti-MRSA activity of JR20.

JR20 exhibited potent inhibitory effects on biofilm formation, as evidenced by the absence of crystal violet staining at concentrations approximating 2 × IC₅₀, suggesting complete suppression of biofilm development (Figure 2A). In contrast, the compound's biofilm disruption capacity was not clearly demonstrated in this initial assay.

SYTO 9 dye typically labels all bacteria within a population—regardless of whether their cell membranes are intact. PI, however, can only penetrate bacteria with damaged membranes. Upon binding to DNA, PI exhibits enhanced fluorescence intensity, displacing SYTO 9 and causing dead cells to appear red instead of green (Stiefel et al., 2015). Thus, bacteria with intact cell membranes exhibit green fluorescence, while those with damaged membranes appear red. The excitation/emission peaks for these dyes are: SYTO 9 at approximately 480/500 nm, and PI at ~490/635 nm (Mcgoverin et al., 2020).

Confocal microscopy was employed to quantitatively assess the inhibitory and disruptive effects of JR20 against MRSA biofilms, using red/green fluorescence intensity and biofilm thickness as key metrics, over a total imaging area of 135 × 135 μm². In the biofilm inhibition assay (Slide 1, Figure 2B), both the positive control vancomycin (VAN) and JR20 at 2 × IC₅₀ and 4 × IC₅₀ effectively suppressed biofilm formation, as reflected by negligible fluorescence signals (non-fluorescent area > 99 %) and minimal thickness (~3 μm) (Figure 2D). At lower concentrations (0.5 × IC₅₀ and 1 × IC₅₀), moderate inhibition was observed, with biofilm formation rates of only 16–18 % relative to the negative control.

In the biofilm disruption assay (Slide 2, Figure 2C), JR20 at 0.5 × IC₅₀ induced partial bacterial death (8.69 %) and reduced biofilm thickness. Higher concentrations (1 × IC₅₀ to 4 × IC₅₀) progressively enhanced damage, marked by further decreases in thickness, a gradual rise in the red-fluorescence ratio from 11.68% to 15.41%, and nearly absent green fluorescence, consistent with widespread loss of membrane integrity. Of note, the total fluorescent area slightly exceeded the sum of the individual red and green areas, indicating the presence of mixed signals or non-specific background fluorescence—an expected feature of this experimental system. The accompanying pie chart illustrates the relative proportions of red (dead cells), green (live cells), and non-fluorescent areas within the total projected area; these values represent normalized distributions, as the measured regional areas did not sum to 100% due to signal overlap and background noise.

Collectively, these concentration-dependent effects demonstrate JR20′s dual capacity to inhibit biofilm formation and disrupt pre-formed biofilms, supporting further investigation into its underlying mechanisms and suggesting potential bactericidal activity against MRSA.

Figure 3 presents scanning and transmission electron microscopy images of MRSA following treatment. In the negative control group, MRSA cells displayed typical morphology with clear division septa, intact cell walls, smooth surfaces, and uniformly turgid cellular structures. In the vancomycin (VAN) positive control group, early signs of cytoplasmic dissolution were observed (red circle), accompanied by ill-defined cellular boundaries and tumor-like aggregated structures under SEM, along with visible biofilm proliferation (yellow arrows).

Treatment with low concentrations of JR20 resulted in cellular collapse and minor cytoplasmic leakage (red arrows). At higher JR20 concentrations, significant tumor-like aggregation, extensive biofilm-like material or cellular debris accumulation (green arrows), and substantial cell wall damage (red box) were evident. These morphological alterations demonstrate that JR20 disrupts MRSA cell wall integrity and interferes with biofilm formation, ultimately leading to bacterial necrosis.

Intracellular ATP levels serve as a key indicator of antibacterial efficacy and provide insights into the drug's mechanism of action. As shown in Figure 4, at low concentrations of JR20 (1 × IC₅₀ to 2 × IC₅₀), ATP levels initially increased relative to the negative control, a trend also observed with 20 μM vancomycin (VAN). This transient rise suggests that MRSA may activate a stress response to counteract the drug by boosting metabolic activity. However, as the JR20 concentration increased from 2 × IC₅₀ to 8 × IC₅₀, ATP levels progressively declined, indicating a shift from transient metabolic activation toward metabolic collapse and cell death. The marked ATP suppression at elevated concentrations likely reflects irreversible cellular damage, potentially mediated through cell wall disruption or inhibition of key enzymatic activities. This concentration-dependent ATP depletion correlates well with the ultrastructural damage visualized by scanning and transmission electron microscopy (SEM/TEM), collectively supporting the conclusion that JR20 induces bacterial necrosis. These converging lines of evidence underscore the need for further investigation into the precise mechanism of action of JR20.

Molecular docking analysis was performed to gain preliminary mechanistic insights into the JR20-PBP2a interaction. Analysis using visualization software (PyMOL and Discovery Studio 2021) indicated that JR20 binds to PBP2a with a high affinity (−7.1 kcal/mol, Figure 5A), which is stronger than that of penicillin G (−5.9 kcal/mol, Figure 5B). Structural observations further revealed that JR20 forms one hydrogen bond with Lys218 and additional carbon-hydrogen bonds with neighboring residues within the non-binding allosteric domain. These interactions appear to collectively contribute to complex stability. Taken together, these results suggest that JR20 may modulate PBP2a function via binding to this allosteric domain.

Previous studies indicate that the α2-α3 and β3-β4 loop regions in PBP2a, shield the active site and impede antibiotic entry into the active pocket. Conformational changes in the β3 chain and the N-terminal region of the α2 helix are essential for β-lactam antibiotics to approach Ser403 (Mahasenan et al., 2017; Rosado et al., 2025). Based on these findings, we propose a bold hypothesis: although JR20 does not occupy the active site of this protein, its binding position suggests a possible allosteric regulatory mechanism. Namely, JR20 binds to the active region within the enzyme's N-terminal non-penicillin-binding domain, inducing an allosteric effect that inhibits cell wall synthesis.

Ly6G, as a neutrophil-specific marker, is commonly used in immunohistochemical (IHC) or immunofluorescence (IF) staining analyses. In H&E staining, cell nuclei appear vivid blue after hematoxylin staining and mucus appears grayish-blue. Cytoplasm exhibits a pink-to-peach gradient after eosin staining, and intracellular eosinophilic granules show bright red reflectivity (Adrover et al., 2025; Feldman and Wolfe, 2014).

Spleen H&E staining revealed: the model group exhibited extensive deep blue inflammatory infiltration with blurred splenic corpuscle borders and prominent red eosinophilic granules. This indicated the activation of inflammation-mediated immune responses (yellow circles). This feature was significantly improved in the JR20-treated group (Figure 6A). Ly6G staining revealed extensive darkly stained neutrophil infiltration around splenic corpuscles in the model group (red arrows). Partial neutrophil infiltration persisted in the low-dose JR20 group, while the high-dose group showed markedly reduced infiltration (Figure 6B).

Liver H&E staining demonstrated fibrotic necrosis in the hepatic parenchyma of the model group (yellow arrows). Fibrotic lesions remained present in the low-dose JR20 group. Whereas, no significant abnormalities were observed in the high-dose group (Figure 6C). Ly6G staining similarly revealed marked neutrophil infiltration in the model group liver (black arrows), with reduced infiltration observed across all treatment groups (Figure 6D). These sections indicate that extensive neutrophil infiltration and inflammatory infiltration characteristics in the model group, collectively suggest a state of profound inflammation. JR20 effectively alleviates this pathological manifestation, with therapeutic effects exhibiting a concentration-dependent pattern.