Curcumin-mediated photodynamic inactivation of Vibrio parahaemolyticus biofilms on stainless steel: Oxidative damage and degradation of exopolysaccharides Authors: Xiulian Huang, Lu Chen, Shiying Tang, Zhiyun Peng, Zihao Lin, Konglong Feng, Langhong Wang, Yong Zhao, Qiaohui Zeng, Jing Jing Wang Published: 2026 ABSTRACT Vibrio parahaemolyticus biofilms on stainless steel surfaces pose a serious threat to food safety. This study investigated the anti-biofilm efficacy and eradication mechanism of curcumin -mediated photodynamic inactivation (PDI) by targeting exopolysaccharides, a key component of extracellular polymeric substances (EPS) within biofilms. Results showed that the PDI treatment with 60 μM curcumin and 9.17 J/cm2 blue light irradiation eradicated V. parahaemolyticus biofilms by >60% and killed 5.0 log CFU/mL cells. Such robust efficiency was attributed to the fact that the PDI treatment greatly cleaved the glycosidic linkages and reduced molecular weight of exopolysaccharides, consequently destroying the structures of hydrogen-bonding networks. These changes decreased the hydrophobicity, reduced the adhesion work and diminished the interfacial tension of EPS matrix, which caused a profound loss of both cohesive and interfacial stability of the biofilm. As a result, the PDI treatment induced the structural disruption and fragmentation of the biofilm network, which contributed to the above high reductions of biomass and sessile cells. This study elucidates the mechanism of PDI against biofilms by targeting the key component of EPS, thereby providing theoretical guidance for the development of potent PDI. Introduction Biofilms formed by foodborne pathogens on food-contact surfaces pose a major threat to food safety and public health. It has been reported that biofilm-related contamination of equipment has been estimated to cause 59% of foodborne disease outbreaks (Midelet and Carpentier, 2004). These structured communities are encased within a self-produced matrix of extracellular polymeric substances (EPS) (Flemming et al., 2023). The EPS matrix constitutes a cohesive, protective three-dimensional scaffold that not only mediates bacterial adhesion to contact surfaces but also shields the embedded cells against a broad spectrum of environmental stresses and antimicrobial agents. Vibrio parahaemolyticus is of particular concern in seafood processing given its halophilic trait and high prevalence in aquatic environments, with detection rates of 35%–70% in seafood products (Wu et al., 2024). Its ability to form robust biofilms on stainless steel produces a persistent cross-contamination risk. Compared with planktonic cells, biofilm formation of V. parahaemolyticus not only promotes bacterial survival under nutrient deprivation, desiccation and low temperatures, but also endows strong tolerance to industrial sanitizers (Sauer et al., 2022). This recalcitrance mainly derives from the complex architecture and physicochemical properties of EPS matrix, which acts as a diffusion barrier to neutralize reactive agents and physically shields embedded cells. Therefore, conventional sanitation protocols (chemical disinfection combined with mechanical cleaning) are largely ineffective in eradicating biofilms (Elafify et al., 2024), which highlights an urgent demand for alternative and efficient anti-biofilm approaches. Photodynamic inactivation (PDI) has emerged as a promising non-thermal antimicrobial technology in the food industry (Chen et al., 2025). Its mechanism relies on non-toxic photosensitizers that, upon excitation by light in the presence of oxygen, generate cytotoxic reactive oxygen species (ROS) including singlet oxygen (1O2), hydroxyl radicals (•OH), etc. These highly reactive molecules induce non-selective oxidative damage to microbial lipids, proteins, and nucleic acids, ultimately causing cell death (Chen et al., 2023). At present, the biofilm-eradicating and bactericidal effects of PDI are well-documented, and notably the curcumin (Cur)-mediated PDI has been shown to significantly reduce the biomass of V. parahaemolyticus biofilms by over 70% and notably decrease the content of exopolysaccharides (Tan et al., 2018). However, the underlying relationships between the ROS-mediated impairment of EPS components and the disintegration of mature biofilms remains unclear. EPS matrix is not a homogeneous amorphous slime but a highly ordered, hydrated composite polymer, whose functional integrity is pivotal for maintaining biofilm stability, intercellular cohesion, and environmental stress resistance. It is primarily composed of exopolysaccharides, proteins, extracellular DNA (eDNA), and lipids, with exopolysaccharides typically accounting for the most abundant fraction (40–95% of the dry mass) (Xu et al., 2024). Serving as the primary three-dimensional structural scaffold of biofilms, high-molecular-weight exopolysaccharides, which comprise repeating monosaccharide units bound by glycosidic linkages, generate a viscous gel-like network via hydrogen bonding and hydrophobic forces, thereby improving the rigidity and structural stability of the EPS matrix (Deo et al., 2025). Accordingly, targeting the exopolysaccharide component represents a strategic and scientifically rational approach for biofilm eradication (Tan et al., 2018). From a molecular targeting perspective, the generated ROS during curcumin-mediated PDI can cleave glycosidic bonds or modify functional groups on polysaccharide side chains, thereby disrupting the delicate balance of intermolecular interactions between exopolysaccharides and other EPS constituents. This disruption may induce polysaccharide depolymerization, trigger composite matrix disassembly, and ultimately compromise the physical barrier and structural integrity of biofilm (Wang et al., 2024). Therefore, exploring the biofilm disruption through the lens of exopolysaccharide-targeted strategies is scientifically well-founded. This study revealed the mechanisms by which curcumin-mediated PDI eradicates mature biofilms of V. parahaemolyticus, focusing on the molecular structure and functions of exopolysaccharides. Firstly, the eradication efficacy of PDI against V. parahaemolyticus biofilms was investigated by monitoring the changes of biofilm biomass, bacterial viability, and three-dimensional structures. Subsequently, the exopolysaccharides were isolated to elucidate the specific changes on their chemical composition, molecular structure, and nanoscale morphology following the PDI treatment, by employing a comprehensive suite of analytical techniques, including gel permeation chromatography (GPC), high-performance liquid chromatography (HPLC), Fourier-transform infrared (FT-IR) spectroscopy, Raman spectroscopy, and atomic force microscopy (AFM). This study offers important technical insights and theoretical support for the green and efficient control of bacterial biofilms in the food industry. Materials and methods Material V. parahaemolyticus (VP17, VP36, VP49, and VP17802) were provided by the laboratory of Shanghai Ocean University. All four strains are clinical isolates originally obtained from fecal samples. Thiosulfate citrate bile salt sucrose agar (TCBS) and tryptone soy broth (TSB) were purchased from Beijing Landbridge Technology Co., Ltd. (Shanghai, China). Crystal violet dye was provided by Shanghai Macklin Biochemical Co., Ltd. MTT assay kit was purchased from Solarbio Life Sciences Co., Ltd. (Beijing, China). PI staining solution was provided by Beyotime Biotechnology Co., Ltd. (Shanghai, China). Phenol, sulfuric acid, anhydrous ethanol, and other analytical reagents were purchased from Sinopharm Chemical Reagent Co., Ltd. Bacterial strains and biofilm formation The bacterial strains were stored at −80 °C. For recovery, bacteria were streaked onto TCBS agar plates and incubated at 37 °C for 12 h. Single colony was selected and inoculated into TSB supplemented with 3% NaCl. For all experiments in this study, a cocktail of the four strains was prepared by mixing equal volumes of each overnight culture to a final concentration of approximately 8.0 log CFU/mL. The rationale for using a cocktail was to account for strain-to-strain variability in biofilm formation and PDI susceptibility, which better reflects the microbial diversity encountered in real food processing environments. The cultures were incubated overnight at 37 °C with shaking at 180 rpm. Finally, bacterial cells were harvested by centrifugation, washed repeatedly with 0.85% NaCl, and resuspended to a concentration of ∼8.0 log CFU/mL for subsequent biofilm formation. Light-emitting diodes (LED) system and photosensitizer Blue light-emitting diodes (LEDs, 455-460 nm, 20 W) were purchased from Getian Opto-Electronics (Shenzhen China). LEDs were placed in a deep photo accessory, and their power density was measured at 5.09 mW/cm2. The energy dosage was calculated by Eq. (1) (Li et al., 2021):(1)E=P×twhere E is energy density (J/cm2), P is power density (W/cm2), and t is time (s). Curcumin with a purity of 95% was purchased from Sangon Biotechnology (Shanghai, China). Curcumin powder was weighed and dissolved in 10 mL of anhydrous ethanol to obtain a curcumin stock solution (1 mM). The solution was then stored at 4 °C in the dark condition. Prior to each experiment, the stock solution was freshly diluted to the working concentration in the culture medium and immediately used without prolonged storage in aqueous solution. The final concentration of ethanol in the culture medium was 0.1% (v/v). Published studies have demonstrated that ethanol at concentrations ≤50% exhibits negligible antibacterial activity (Wu et al., 2024; Prochnow et al., 2016); therefore, the 0.1% ethanol present in curcumin-treated samples does not contribute to the observed anti-biofilm effects. Formation and quantification of biofilms For biofilm formation, 990 μL of TSB with 3% NaCl was mixed with 10 μL of V. parahaemolyticus suspension in a 24-well plate. A sterile stainless steel coupon (13 mm diameter × 0.2 mm thickness) was placed into each well, and the plate was statically incubated at 25 °C for 12, 24, 36, 48, and 60 h to allow biofilm formation on the coupon surfaces. After incubation, planktonic cells were removed, and the biofilm was gently rinsed with phosphate-buffered saline (PBS). The coupons were then transferred to a new 24-well plate and dried at 55 °C, and 1 mL of 0.1% (v/v) crystal violet solution was added to each well containing a coupon, followed by incubation for 30 min. Excess dye was removed by PBS. Then, 1 mL of 95% ethanol was added to each well and incubated for 30 min to solubilize the bound crystal violet. Biofilm formation was quantified by measuring the absorbance at 600 nm. It should be noted that crystal violet staining is not specific to bacterial cells; it binds to both cells and extracellular polymeric substances. Therefore, the OD600 measurement reflects total biofilm biomass (cells + matrix) rather than cell count alone. To specifically assess cell viability, complementary methods (colony counting, MTT assay, and PI staining) were used as described below. Eradication procedure of PDI against biofilms To evaluate the eradication efficacy of PDI against V. parahaemolyticus biofilms, mature biofilms formed on stainless steel were incubated with serial concentrations of curcumin (0, 20, 40, 60, 80 μM) for 20 min. The plates were then exposed to light for 10 min (3.05 J/cm2), 20 min (6.11 J/cm2), 30 min (9.17 J/cm2), 40 min (12.22 J/cm2). The treated biofilms were then homogenized to release sessile cells from the stainless steel coupons, and serial dilutions were plated on TCBS agar plates. After incubation at 37 °C for 36 h, CFU were counted to assess the antimicrobial efficacy of PDI. Results are expressed as log CFU per mL of homogenized biofilm suspension. Two types of controls were established: the samples without light and curcumin (L-C-) were defined as blank control, while the samples treated with curcumin alone in the dark (L-C+) were selected as curcumin control. Following PDI treatment, the biofilms were mixed with 1 mL of fresh TSB medium and 100 μL of MTT reagent. The plates were then incubated statically at 25 °C for over 4 h. After removing planktonic cells, 1 mL of dimethyl sulfoxide (DMSO) was added to each well. The plates were left for 30 min, after which the absorbance was measured at a wavelength of 570 nm (Han et al., 2017). To visualize bacterial death in biofilms, propidium iodide (PI) staining was performed. Biofilm samples were stained with PI and incubated in the dark for 30 min. The distribution of dead and viable cells within the biofilms was examined via confocal laser scanning microscopy (CLSM). Extraction of extracellular polymeric substance (EPS) The biofilm samples were collected and subjected to ultrasonic disruption (47.5 W, 5 s on/5 s off, 9 cycles) using a cell disruptor to separate bacterial cells from the EPS matrix. Subsequently, the resulting suspension was centrifuged at 8000 rpm for 5 min at 4 °C to obtain the supernatant. The supernatant was then filtered through a 0.22 μm membrane for subsequent analysis. Changes in structural parameters of biofilms The biofilms were stained with 4S Green fluorescent dye. Afterwards, the 24-well plate was incubated in the dark for 30 min. The structural changes were observed using CLSM, and three random fields of view were selected and imaged under a 20× objective lens. Acquired images were analyzed using ZEM image acquisition software and the open-source ImageJ software to quantify the changes of structural parameters, average fluorescence intensity, and biofilm porosity (Tan et al. 2022). Determination of contact angle and interfacial tension Following the method of Sauer et al. (2022), the contact angle of EPS solution (100 μg/mL) was measured using a contact angle goniometer. A 2 μL of droplet was deposited onto hydrophobic aluminum foil using the instrument's automatic dispensing system. Subsequently, a high-magnification camera captured the droplet profile, which was used to determine the contact angles. The surface tension of EPS solution was measured by the interfacial tension mode. The adhesion work of EPS solution was then calculated based on Young's equation:(2)γsg=γsl+γlgcosθ The surface tension of EPS solution at the oil-water interface was measured as a function of time by the interfacial tension mode. A pendant drop was automatically formed and suspended at the oil-water interface in a transparent glass cuvette containing sunflower oil. Interfacial tension changes were recorded at 1-min intervals over 90 min. Quantification of exopolysaccharides and biofilm biomass after PDI treatment The exopolysaccharide was quantified using phenol-sulfuric acid method. Briefly, an aliquot of the supernatant was mixed with 6% phenol and concentrated sulfuric acid. The mixture was vortexed thoroughly and allowed to stand for 30 min. Absorbance was then measured at 490 nm using a microplate reader. A glucose standard curve was generated to calculate EPS content, expressed as glucose equivalents (μg/mL of extract). Subsequently, total biomass was quantified in an independent set of PDI-treated samples via crystal violet staining. Extraction of exopolysaccharides Exopolysaccharides were extracted following the previous study of Wang et al. (2024). Mature biofilms were enriched in centrifuge tubes and then centrifuged at 10,000×g for 15 min to collect the supernatant. 30 mL of ethanol was added at a supernatant-to-ethanol volume ratio of 3:1, and the mixture was stirred at 4 °C for 4 h. The mixture was centrifuged to obtain the precipitate, which was dissolved in 9 mL of Tris-HCl buffer (pH 8.0, 0.05 M). Proteinase K was added to a final concentration of 0.2 mg/mL, and the mixture was incubated at 37 °C for 4 h, followed by heating in a 95 °C water bath for 10 min. Another 30 mL of ethanol was added, and the mixture was stirred at 4 °C for 4 h. After centrifugation, the resulting precipitate was dissolved in pure water, dialyzed for 48 h, and lyophilized. Characterization of exopolysaccharides Molecular weight (MW) and monosaccharide composition analysis The molecular weight (Mw) of exopolysaccharides was measured by Gel Permeation Chromatography (GPC, Agilent 1260, USA), with a TSK-Gel G4000PWXL column (7.8 × 300 mm). The column was eluted with ultrapure water at 1 mL/min (35 °C), calibrated with dextran standards. The monosaccharide composition was analyzed using high-performance liquid chromatography (HPLC) after TFA hydrolysis. Identification and quantification of monosaccharides were conducted based on retention time and external standard calibration curves. Standards included fucose (Fuc), arabinose (Ara), rhamnose (Rha), galactose (Gal), glucose (Glc), xylose (Xyl), mannose (Man), fructose (Fru), ribose (Rib) and their corresponding uronic acids. Fourier transform infrared (FT-IR) and Raman spectroscopy Changes in the chemical structure of PDI-treated exopolysaccharides were determined by FT-IR and Raman spectroscopy. In detail, 1 mg of exopolysaccharides and 200 mg KBr were mixed, ground in a mortar, and an appropriate amount was compressed into pellets. FT-IR spectra were recorded using an Excalibur 3100 spectrometer (Varian, USA) over the wavenumber range of 4000–400 cm−1. Exopolysaccharides were placed on aluminum slides, and Raman spectra were acquired for each sample with an acquisition time of 20 s using a spectrometer (Agiltron, USA) equipped with a 785 nm diode laser. The laser power was set to 40 mW, and the laser was focused on the sample through a 10× objective lens. AFM imaging analysis The nanostructures of exopolysaccharides were characterized by atomic force microscopy (AFM). Diluted exopolysaccharide solution was deposited on a freshly cleaved mica substrate and air-dried before imaging. Measurements were performed in tapping mode using silicon cantilever, and at least three distinct areas per sample were scanned. The acquired height images were processed and analyzed via Nanoscope Analysis software for flattening and surface roughness determination. Statistical analysis All experiments results were repeated independently three times, and each trail contained at least three samples. Experimental outcomes were reported as mean values ± standard deviation. One-way analysis of variance was conducted (P < 0.05). Data analysis and statistical computations were performed using SPSS Statistics 21.0 (SPSS Ink., Armonk, NY), following the standard procedures for hypothesis testing in quantitative analysis. Results and discussion Eradication efficiency of curcumin-mediated PDI against V. parahaemolyticus biofilms Biofilm growth on stainless steel coupons was characterized by crystal violet staining for biomass quantification (Fig. 1A). The growth curve revealed a temporal trend where the biomass initially increased and then decreased, peaking at 36 h and declining by 48 h. The biomass typically accumulates until maturation, after which partial cellular detachment or structural disintegration occurs, leading to reduced values (Zhang et al.,2023). Accordingly, the mature biofilms at 36 h were selected for subsequent experiments.Fig. 1Formation of V. parahaemolyticus Biofilms and the effects of the curcumin-mediated PDI on biofilm biomass and cell viability. Biofilm formation (A). Quantification of biofilm biomass (B) and the effects of curcumin concentrations (0, 20, 40, 60, 80 μM) on the cell viability (C) following the PDI treatment. (D) The effects of irradiation time (0, 10, 20, 30, 40 min, corresponding to 0, 3.05, 6.11, 9.17, 12.22 J/cm2) on cell viability following the PDI treatment. (E) Changes in the metabolic activity of cells after PDI treatment. (E) Changes in the metabolic activity of cells after the PDI treatment. (F) Fluorescence microscopy images of PI-stained cells within biofilms: (F1) blank control (L-C-); (F2) curcumin control (L-C+); (F3-F5) PDI treatment with 40, 60, and 80 μM curcumin, respectively, and 9.17 J/cm2 irradiation. Letters (a-f) represent the significant difference (P < 0.05) of the values in (A), (B), (C), (D), and (E). Formation of V. parahaemolyticus Biofilms and the effects of the curcumin-mediated PDI on biofilm biomass and cell viability. Biofilm formation (A). Quantification of biofilm biomass (B) and the effects of curcumin concentrations (0, 20, 40, 60, 80 μM) on the cell viability (C) following the PDI treatment. (D) The effects of irradiation time (0, 10, 20, 30, 40 min, corresponding to 0, 3.05, 6.11, 9.17, 12.22 J/cm2) on cell viability following the PDI treatment. (E) Changes in the metabolic activity of cells after PDI treatment. (E) Changes in the metabolic activity of cells after the PDI treatment. (F) Fluorescence microscopy images of PI-stained cells within biofilms: (F1) blank control (L-C-); (F2) curcumin control (L-C+); (F3-F5) PDI treatment with 40, 60, and 80 μM curcumin, respectively, and 9.17 J/cm2 irradiation. Letters (a-f) represent the significant difference (P < 0.05) of the values in (A), (B), (C), (D), and (E). The anti-biofilm efficacy of curcumin-mediated PDI was evaluated using crystal violet staining and colony counting methods. In Fig. 1B and C, the curcumin treatment (60 μM) or light irradiation (9.17 J/cm2) did not significantly reduce biofilm biomass (P > 0.05). This indicated that single treatment did not eradicate the biofilms, which was consistent with our previous findings. Remarkably, the PDI treatment with 20 μM curcumin and 30 min of light exposure (9.17 J/cm2) decreased the OD600 of biofilm biomass to 0.71, resulting in a significant (P < 0.05) decrease relative to the control. When the curcumin was increased to 60 μM, the PDI exhibited an enhanced ability to eradicate the biofilms with OD600 decreasing to 0.36. However, a further elevation of curcumin concentration to 80 μM resulted in a slight rebound in biofilm biomass. This phenomenon may be attributed to aggregation-caused quenching (ACQ) of curcumin, which consequently compromised its photodynamic activity (Chen et al., 2025). When the curcumin was fixed at 60 μM, the PDI with 3.05 J/cm2 (10 min) irradiation reduced the viable cells from an initial 7.6 log CFU/mL to 7.2 log CFU/mL (Fig. 1D). When the light dose was increased to 9.17 J/cm2 (30 min), the PDI reduced the cells to 2.6 log CFU/mL, and a further increase of light dose (12.22 J/cm2) did not increase the inactivation efficiency. The observed dose-dependent inactivation confirmed a direct correlation between light energy and bactericidal efficacy of PDI. Based on these results, the combination of 60 μM curcumin and 9.17 J/cm2 irradiation constitutes an effective PDI regimen, which not only achieves great reductions in both biofilm biomass and viable cells, but also avoids the ACQ effect of curcumin. The MTT assay was employed to further quantify the metabolic activity of cells within biofilms, serving as a functional indicator of cellular viability (Koca and Taslimi, 2023). In Fig. 1E, the blank control exhibited an OD570 value of 1.37, indicative of robust metabolic activity. In contrast, the biofilms treated with 60 μM curcumin and 9.17 J/cm2 irradiation showed a pronounced reduction of OD570 to 0.28, corresponding to an approximate 79.6% reduction. This phenomenon suggests that the curcumin-mediated PDI decreases the metabolic activity of cells within biofilms. PI penetrates the cell membrane of dead cells with compromised membrane integrity, intercalates with DNA, and emits red fluorescence (Shi et al., 2022). In Fig. 1F, the visual analysis confirmed the eradication efficacy of curcumin-mediated PDI against biofilms. Compared with control (F1), single curcumin (60 μM) treatment (F2) yielded almost no red fluorescence signal, indicating negligible residual biomass. In contrast, the PDI-treatment groups (F3-F5) exhibited a gradual increase in red fluorescence with the addition of curcumin. Notably, the PDI treatment with 60 μM curcumin and 30 min irradiation displayed the most extensive red fluorescence. These results collectively validate the synergistic anti-biofilm effect of curcumin and light irradiation under the action of PDI. It should be noted that the optimal PDI parameters are highly system-dependent. Under our experimental conditions (stainless steel coupons, V. parahaemolyticus mixed-strain biofilm), 60 μM curcumin and 30 min irradiationachieved maximum biofilm eradication. The reduced efficacy at 80 μM is attributed to ACQ of curcumin, which compromises ROS generation at supra-optimal concentrations (Chen et al., 2025). This concentration-dependent pattern is consistent with other studies. For instance, Tian et al. (2026) reported optimal inactivation of Burkholderia gladioli using 144 μM curcumin with 21 min illumination, while Wang et al. (2023) achieved significant antifungal effect against Penicillium italicum using 75 μM curcumin with 20 min treatment. These variations reflect differences in target microorganisms, biofilm maturation, substrate properties, and light sources, rather than inconsistencies. Changes in the structural and functional properties of extracellular polymeric substance (EPS) CLSM was employed to evaluate the structural changes in biofilms on stainless steel surfaces. As shown in Fig. 2A, the blank control exhibited an irregular surface with dense and heterogeneous protrusions, which was a hallmark of robust EPS accumulation conferring mechanical stability. Such phenomenon was consistent with the cross-linked polysaccharide-protein networks reported previously (Zhang et al. 2022). The curcumin control did not alter the roughness of biofilms. In contrast, the PDI treatment with a light dose of 9.17 J/cm2 led to a progressive increase in the surface smoothness with the increase of curcumin. Especially with 60 μM curcumin, the biofilm surfaces became very smooth, indicating that the generated ROS degraded the EPS components and triggered the structural collapse of biofilms. When the curcumin concentration was increased to 80 μM, significant changes were not observed, owing to the ACQ effects of high curcumin concentration (Tan et al., 2021).Fig. 2Effects of the curcumin-mediated PDI on the changes of textural structures, contact angles, adhesion work, interfacial tension of EPS matrix. (A) CLSM images of biofilms on stainless steel. (B) 3D images of biofilms. (C) Porosity and contact angles of EPS matrix. (E) EPS adhesion work. (F) EPS interfacial tension. (A1, B1) untreated control; (A2, B2) curcumin alone (60 μM); (A3-A5, B3-B5) PDI treatment with 40, 60, 80 μM curcumin and blue light (9.17 J/cm2). Letters (a-d) represent the significant difference (P < 0.05) of the values in (C), (E). Effects of the curcumin-mediated PDI on the changes of textural structures, contact angles, adhesion work, interfacial tension of EPS matrix. (A) CLSM images of biofilms on stainless steel. (B) 3D images of biofilms. (C) Porosity and contact angles of EPS matrix. (E) EPS adhesion work. (F) EPS interfacial tension. (A1, B1) untreated control; (A2, B2) curcumin alone (60 μM); (A3-A5, B3-B5) PDI treatment with 40, 60, 80 μM curcumin and blue light (9.17 J/cm2). Letters (a-d) represent the significant difference (P < 0.05) of the values in (C), (E). The curcumin control exhibited nearly identical fluorescence patterns to the blank control (Figs. 2B–1), confirming that curcumin alone did not affect the morphological structures of biofilms (Fig. 2B–). By contrast, the curcumin-mediated PDI caused a significant widening of interstices within the biofilms and a notable reduction in fluorescence intensity. Notably, the PDI with 60 μM curcumin and 9.17 J/cm2 irradiation (Fig. 2BFig. 4-4 and 2.5D) reduced the fluorescence intensity to undetectable, suggesting that the architectures of biofilms were destroyed.Fig. 3Changes of (A) exopolysaccharides and (B) biomass of V. parahaemolyticus biofilms after PDI treatment. (C) The correlation analysis between the exopolysaccharides and biofilm formation. (D) The absorbance value of the EPS suspension and the purified exopolysaccharides extracted from V. parahaemolyticus biofilms. (E) SDS-PAGE bands and (F) agarose gel of the EPS suspension and the purified exopolysaccharides, where T represents the purified exopolysaccharides. (G) Biomass and (H) viability of sessile cells. Letters (a-d) represent the significant difference (P < 0.05) of the values in (A), (B).Fig. 4Structural analysis of exopolysaccharides following PDI treatment. (A) HPLC analysis of monosaccharide composition. (B) FTIR spectra of functional groups. (C) Raman spectra of molecular fingerprints. (D) AFM height images: (D1) blank control (L-C-); (D2) curcumin control (L-C+); (D3) PDI-treated (60 μM curcumin and 9.17 J/cm2 irradiation). Scale bars in (D1–D3) correspond to 30 μm; the Z-scale is consistent across all AFM images. Changes of (A) exopolysaccharides and (B) biomass of V. parahaemolyticus biofilms after PDI treatment. (C) The correlation analysis between the exopolysaccharides and biofilm formation. (D) The absorbance value of the EPS suspension and the purified exopolysaccharides extracted from V. parahaemolyticus biofilms. (E) SDS-PAGE bands and (F) agarose gel of the EPS suspension and the purified exopolysaccharides, where T represents the purified exopolysaccharides. (G) Biomass and (H) viability of sessile cells. Letters (a-d) represent the significant difference (P < 0.05) of the values in (A), (B). Structural analysis of exopolysaccharides following PDI treatment. (A) HPLC analysis of monosaccharide composition. (B) FTIR spectra of functional groups. (C) Raman spectra of molecular fingerprints. (D) AFM height images: (D1) blank control (L-C-); (D2) curcumin control (L-C+); (D3) PDI-treated (60 μM curcumin and 9.17 J/cm2 irradiation). Scale bars in (D1–D3) correspond to 30 μm; the Z-scale is consistent across all AFM images. ImageJ analysis further corroborated the structural disintegration induced by PDI. In Fig. 2C, the biofilm porosity, an indicator of structural compactness, exhibited a dose-dependent increase following PDI treatment. Notably, the PDI treatment with 60 μM curcumin and 9.17 J/cm2 irradiation increased the porosity to 83%, which indicated that a profound disruption of the EPS matrix and the formation of a highly porous, loose textural structure in the biofilms. Consequently, the PDI treatment impaired the structural integrity of biofilms and abrogated their adhesive stability. The adhesive capacity of EPS matrix is closely associated with their hydrophilicity/hydrophobicity, with the surface free energy being a key determinant of interfacial adhesion. Previous studies have indicated that hydrophobic interactions are critical forces sustaining the three-dimensional network structures of biofilms (Sylwia,2010), and generally biofilms with lower surface free energy are more prone to adhere to contact surfaces. Fig. 2D showed that there was no significant difference in the contact angles between blank control (58.68°) and curcumin control (58.46°). In contrast, the curcumin-mediated PDI significantly elevated the contact angles from 75.04° to 84.27°, which indicated a marked reduction in the hydrophobicity of EPS matrix. The depletion of hydrophobic components within the EPS matrix weakened the internal cohesion of biofilms, thereby enhancing the susceptibility of biofilms to physical and chemical disruption (Flemming et al., 2023). Young's equation was further employed to analyze changes in the adhesion work of EPS after PDI treatment. Adhesion work refers to the energy required for a liquid to spread on a contact surface, and generally, a decrease in adhesion work reflects impaired liquid spreading on the surface, thereby corresponding to diminished adhesive capacity (Liu and Zhao, 2005). In Fig. 2E, the blank control (111.7 J/m2) did not show significant effect on the adhesion work of EPS compared to the curcumin control (110.9 J/m2) (P > 0.05). In contrast, the PDI treatment greatly decreased the adhesion work of EPS from 97.1 J/m2 to 89.8 J/m2, indicating that both the hydrophobicity and adhesive capacity of EPS matrix were greatly reduced. The pendant drop method was employed to evaluate the effects of PDI treatment on the interfacial behavior of EPS. In Fig. 2F, interfacial tension exhibited a time-dependent decrease across all groups. The blank control and the curcumin control displayed comparable initial interfacial tension values (≈27.6 mN/m). In contrast, the PDI-treated group showed a significantly lower initial interfacial tension of 15.1 mN/m, which remained the lowest throughout the entire monitoring period. In general, a lower interfacial tension directly reflects higher interfacial activity, which indicates that surface-active components adsorb more rapidly and effectively at the liquid–solid interface (Ravera et al., 2021). Therefore, the PDI-induced degradation of EPS matrix fundamentally altered its physicochemical properties, as evidenced by the increased contact angles, reduced adhesion work, and diminished interfacial tension. These concurrent physicochemical changes collectively validate the structural disruption and fragmentation of the biofilm network, leading to a profound loss of both cohesive and interfacial stability of the biofilm. PDI-induced alterations in structures and functions of exopolysaccharides Exopolysaccharides are a pivotal structural component regulating the architecture and formation of biofilms (Wan et al., 2024), and hence, the intrinsic relationship between the reduction of biofilm biomass and exopolysaccharides content was investigated in response to the PDI treatment. In Fig. 3A and B, the PDI treatment simultaneously reduced the biofilm biomass and exopolysaccharides content, showing a dose-dependent relationship with the increase of curcumin. Notably, the PDI treatment with 60 μM curcumin and 9.17 J/cm2 irradiation decreased the biofilm biomass and exopolysaccharides by 60.71% and 66.66%, respectively. In Fig. 3C, further analysis of the relationship between the changes of biofilm biomass and exopolysaccharides content revealed a strong positive linear relationship (R2 = 0.999, P < 0.001), which supported the well-established role of exopolysaccharides as the structural scaffold of EPS matrix (Sutherland, 2001). Likewise, Ma et al. (2022) reported that disruption of exopolysaccharides biosynthesis in P. aeruginosa reduced biofilm biomass, decreased thickness, and increased surface roughness, which aligns closely with the structural disintegration of biofilms following the PDI treatment. Exopolysaccharides were purified to characterize the PDI-induced structural changes, and furthermore, biological enzymes were employed to degrade the bound proteins and nucleic acids. In Fig. 3D–F, the UV-Vis absorption spectrum and gel electrophoresis of the purified fraction did not exhibit characteristic peaks and bands for nucleic acids (∼260 nm) or proteins (∼280 nm), proving that they were completely removed. Furthermore, a positive reaction in the phenol-sulfuric acid assay confirmed the presence of carbohydrates. As quantified in Table 2, the purified fraction from the control group contained 249.71 μg/mg of total sugars, demonstrating the presence of exopolysaccharides. A reverse supplementation assay was performed to functionally validate exopolysaccharides as the key molecular target of PDI. In Fig. 3G and H, the addition of exopolysaccharides (0–100 μg/mL) dose-dependently attenuated the eradication efficacy of PDI against biofilms, accompanied by a corresponding increase in both biofilm biomass and viable cells, in comparison with those samples without exopolysaccharides. This observation was attributed to that exogenous exopolysaccharides competitively scavenged ROS, thereby protecting the endogenous EPS matrix and cells from oxidative damage (Böhning et al., 2024). The structural integrity of the exopolysaccharides was evaluated, and the gel permeation chromatography (GPC) revealed significant changes in the molecular weight profile of exopolysaccharides (Table 1). In detail, the control group exhibited a broad molecular weight distribution (polydispersity index = 8.48). The PDI treatment degraded the molecular structures of exopolysaccharides, characterized by a sharp decrease in the ultra-high molecular weight fraction with Z-average molar mass (Mz) dropping from ∼22,481 kDa to ∼9916 kDa) and a concurrent increase in the number-average (Mn) and weight-average (Mw) molecular weights. This phenomenon indicates that ROS preferentially cleaved the larger polymer chains, leading to fragmentation and narrowed molecular weight distribution (Wan et al., 2024). On this basis, the concomitant elevation of Mn and Mw can be explained by two complementary mechanisms: First, the accumulation of intermediate-sized fragments from ultra-high molecular weight components increases the average molecular weight (Li et al., 2024); Second, oxidative stress may also trigger cross-linking reactions between polysaccharide chains, a well-documented microbial stress response that increases the average molecular weight of exopolysaccharides (Kaur and Dey, 2023).Table 1Molecular weight (kDa) analysis of exopolysaccharides before and after PDI treatment.Table 1Molecular weightControl groupPDI-treatment groupNumber-average molecular weight (Mn)150.756269.838Weight-average molecular weight (Mw)1279.0521697.334average molar mass (Mz)22480.7529915.673Polydispersity (Mw/Mn)8.4846.29Table 2Comparison of monosaccharide composition and total sugar content of exopolysaccharides before and after PDI treatment.Table 2Monosaccharide componentsControl groupPDI-treated groupChange trendTotal sugar content (μg/mg)249.71422.61Significant increase↑neutral sugarsMan49.88 μg/mg (19.98%)70.72 μg/mg (16.73%)Content↑, Proportion↓Glc25.46 μg/mg (10.20%)47.35 μg/mg (11.20%)Content↑, Proportion↑Gal16.70 μg/mg (6.69%)36.10 μg/mg (8.54%)Content↑, Proportion ↑Fuc3.43 μg/mg (1.38%)7.64 μg/mg (1.81%)Content↑, Proportion↑Amino sugarsGalN40.68 μg/mg (16.29%)51.67 μg/mg (12.23%)Content↑, Proportion↓GlcN38.34 μg/mg (15.35%)49.34 μg/mg (11.68%)Content↑, Proportion↓Uronic acidsGlc-UA55.58 μg/mg (22.26%)88.19 μg/mg (20.87%)Content↑, Proportion↓ Table 1Molecular weight (kDa) analysis of exopolysaccharides before and after PDI treatment.Table 1Molecular weightControl groupPDI-treatment groupNumber-average molecular weight (Mn)150.756269.838Weight-average molecular weight (Mw)1279.0521697.334average molar mass (Mz)22480.7529915.673Polydispersity (Mw/Mn)8.4846.29 Molecular weight (kDa) analysis of exopolysaccharides before and after PDI treatment. Table 2Comparison of monosaccharide composition and total sugar content of exopolysaccharides before and after PDI treatment.Table 2Monosaccharide componentsControl groupPDI-treated groupChange trendTotal sugar content (μg/mg)249.71422.61Significant increase↑neutral sugarsMan49.88 μg/mg (19.98%)70.72 μg/mg (16.73%)Content↑, Proportion↓Glc25.46 μg/mg (10.20%)47.35 μg/mg (11.20%)Content↑, Proportion↑Gal16.70 μg/mg (6.69%)36.10 μg/mg (8.54%)Content↑, Proportion ↑Fuc3.43 μg/mg (1.38%)7.64 μg/mg (1.81%)Content↑, Proportion↑Amino sugarsGalN40.68 μg/mg (16.29%)51.67 μg/mg (12.23%)Content↑, Proportion↓GlcN38.34 μg/mg (15.35%)49.34 μg/mg (11.68%)Content↑, Proportion↓Uronic acidsGlc-UA55.58 μg/mg (22.26%)88.19 μg/mg (20.87%)Content↑, Proportion↓ Comparison of monosaccharide composition and total sugar content of exopolysaccharides before and after PDI treatment. Composition analysis of the purified exopolysaccharides revealed a complex monosaccharide profile comprising neutral sugars (Fuc, Ara, Rha, Gal, Glc, Xyl, Man, Fru, Rib) and uronic acids (galacturonic acid (GalUA), guluronic acid (GulUA), glucuronic acid (GlcUA), mannuronic acid (ManUA) (Fig. 4A). Quantitative analysis demonstrated that the PDI treatment substantially increased the total sugar content of exopolysaccharides from 249.71 μg/mg to 422.61 μg/mg, corresponding to an approximate 69% rise (Table 2). The increase is attributed to ROS-mediated cleavage of high-molecular-weight exopolysaccharides into small fragments, which leads to enhanced detectability by the phenol-sulfuric acid method, rather than de novo biosynthesis by surviving cells (Andrew and Jayaraman, 2020). The increase in total sugar content likely represented a microbial stress response, where bacteria attempted to reinforce the matrix by up-regulating polysaccharide synthesis (Vaidya et al., 2025). However, this quantitative increase did not equate to functional integrity, because the concurrent decrease in the molar proportions of key structural sugars (Man, amino sugars, GlcUA) and the cleavage of glycosidic bonds (evidenced by FT-IR/Raman) indicated that the polysaccharides were structurally compromised and lose their scaffolding function (Ma et al., 2022). Although the absolute contents of most neutral sugars (Glc, Gal and Fuc) and the uronic acids GlcUA markedly increased, their relative molar proportions within the total sugar pool exhibited divergent trends. Notably, the absolute content of Glc increased by 86% (Table 3), accompanied by a slight rise in its molar proportion. In contrast, the key structural monosaccharides, including Man, the amino sugars GalN and GlcN, as well as the uronic acid GlcUA, exhibited elevated absolute abundances but reduced relative molar percentages. This shift suggests that the PDI treatment does not merely trigger a global up-regulation of exopolysaccharides synthesis, but rather redirects biosynthetic flux toward specific monosaccharide moieties. This redirection potentially impairs the structural and adhesive functionalities mediated by Man, amino sugars, and uronic acids, a class of components well-recognized for their key roles in sustaining mechanical stability and matrix cohesion of biofilms (Colvin et al., 2012).Table 3Monosaccharide composition analysis (mole percent).Table 3Monosaccharide compositionControl group (mol%)PDI-treatment group (mol%)Glc-UA20.6118.99Man19.9316.41GalN16.3512.06Glc10.1711.51Rib7.715.29Gal6.678.38Fuc1.511.95 Table 3Monosaccharide composition analysis (mole percent).Table 3Monosaccharide compositionControl group (mol%)PDI-treatment group (mol%)Glc-UA20.6118.99Man19.9316.41GalN16.3512.06Glc10.1711.51Rib7.715.29Gal6.678.38Fuc1.511.95 Monosaccharide composition analysis (mole percent). AFM was utilized to probe the nanoscale alterations in exopolysaccharide morphology following PDI treatment (Muller et al., 2021). The blank control presented a continuous and dense network of exopolysaccharides characterized by a rough surface with prominent surface protrusions and aggregates (Figs. 4D–1). The surface topography of biofilms treated with 60 μM curcumin alone (Figs. 4D–2) closely resembled that of control, confirming that curcumin itself does not substantially disrupt the exopolysaccharides scaffold. In contrast, the PDI treatment (Figs. 4D–3) destroyed the continuous and dense network, resulting in the formation of fragmented, scattered, and isolated clusters. This transformation from cohesive architecture to dispersed fragments provided direct visual evidence that the PDI induces the structural disintegration of exopolysaccharide network. (Dufrêne et al., 2021). FT-IR and Raman spectroscopy were further employed to analyze the structural changes within the exopolysaccharides. FT-IR spectrum showed multiple characteristic polysaccharide peaks (Fig. 4B). A noticeable weakening of O-H stretching band around 3424 cm−1 indicated the disruption of hydrogen-bonding network, which is essential for the integrity and hydration of EPS matrix. This alteration aligns with oxidative damage of ROS to hydroxyl groups, which impairs polymer chain cohesion (Zhao et al., 2025). Meanwhile, the enhanced C-H stretching vibrations within the 2800-3000 cm−1 region, corresponding to symmetric and asymmetric modes of methyl and methylene groups, suggested an altered conformational state of exopolysaccharides. Moreover, the absorption peak at 1079 cm−1 exhibited a marked reduction in the intensity accompanied by a discernible wavenumber shift. This peak is primarily associated with the stretching vibration of glycosidic linkage (C-O-C) and coupled C-O stretching vibrations of the pyranose ring backbone (Zhang et al., 2025). Further evidence was observed in the alterations of conformation-sensitive absorption bands. The absorption feature near 1650 cm−1, corresponding to the stretching of carbonyl (C=O) in uronic acids or oxidized sugar rings, showed significant attenuation. Compared with the native peak at 910 cm−1 in control samples, the PDI-treated samples exhibited a new peak at 822 cm−1, which was indicative of residual α-linkages. This progressive loss of anomeric signature vibrations indicates a scission of the glycosidic backbone after the PDI treatment, which compromises the structural integrity of EPS matrix (Liu et al., 2022). Raman spectroscopy provided complementary insights into the vibrational signatures of chemical bonds and is effective in characterizing exopolysaccharides-specific C-O-C linkages (Tahir et al., 2020). In Fig. 4C, the characteristic peak at 1160 cm−1, which is associated with C-O-C glycosidic bond vibrations in the control sample, exhibited significant attenuation following PDI treatment. This pronounced change aligns with the diminished C-O-C/C-O stretching vibration signal observed at 1079 cm−1 in the FT-IR spectrum. These results corroborate that the PDI treatment effectively cleaves the glycosidic linkages within exopolysaccharides backbone, and structurally compromises the EPS matrix. These results demonstrate that the curcumin-mediated PDI disrupts the structural integrity of exopolysaccharides through multi-level oxidative damages. Collectively, the curcumin-mediated PDI induces oxidative damage to the exopolysaccharide network, including cleaving glycosidic linkages, disrupting hydrogen-bonding networks, and altering monosaccharide composition, which lead to the structural fragmentation and impair physicochemical properties of the EPS matrix. These molecular and structural alterations ultimately contribute to the disintegration of biofilms, as well as a cascade of extensive death of sessile cell. Conclusions This study demonstrates that the curcumin-mediated photodynamic inactivation (PDI) effectively eradicates Vibrio parahaemolyticus biofilms on stainless steel surfaces. The anti-biofilm activity arises from a hierarchical cascade triggered by ROS generated upon light irradiation. ROS initially targets the exopolysaccharide framework within the EPS matrix, resulting in the cleavage of glycosidic bonds, disruption of hydrogen-bonding networks, and alteration of monosaccharide profiles. These molecular modifications disrupt the structural backbone of the biofilm matrix, thereby impairing its physicochemical integrity, as reflected by the reduced surface hydrophobicity, weakened adhesion energy, and compromised interfacial stability. Consequently, the biofilm architecture undergoes disintegration, characterized by elevated porosity and loosened structural organization, ultimately leading to widespread bacterial death. It should be noted that this study was performed on clean stainless steel coupons. Collectively, the curcumin-mediated PDI represents a promising non-thermal and eco-friendly strategy for biofilm control in the food industry. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.