Carboxylated graphene quantum dots as a nano-delivery system for insoluble curcumin in antimicrobial photodynamic therapy

Carboxylated graphene quantum dots as a nano-delivery system for insoluble curcumin in antimicrobial photodynamic therapy Kateˇ rina Bartoˇ n Tom´ ankov´ a a,* , Hanna Dilenko a,* , Renata Veˇ ceˇ rov´ a b , Ad´ ela Han´ akov´ a a , Lucie V´ alkov´ a a , Mark´ eta Kolaˇ ríkov´ a a , Luk´ aˇ s Malina a , Robert Bajgar a , Sergii Kalytchuk c,d , Kl´ ara ˇ Cepe c , S.M. Hossein Hejazi c,d , Milan Kol´ aˇ r b , Hana Kol´ aˇ rov´ a a a Department of Medical Biophysics, Faculty of Medicine and Dentistry, Palacký University Olomouc, Hnˇ evotínsk´ a 3, Olomouc 779 00, Czech Republic b Department of Microbiology, Faculty of Medicine and Dentistry, Palacký University Olomouc, Hnˇ evotínsk´ a 3, Olomouc 779 00, Czech Republic c Regional Centre of Advanced Technologies and Materials, Czech Advanced Technology and Research Institute (CATRIN), Palacký University Olomouc, ˇ Slechtitelů 241/ 27, Olomouc 783 71, Czech Republic d Nanotechnology Centre, Centre for Energy and Environmental Technologies, VSB–Technical University of Ostrava, 17. listopadu 2172/15, Ostrava 708 00, Czech Republic ARTICLE INFO Keywords: Antimicrobial photodynamic therapy Curcumin Graphene quantum dots Nanocomposite Antimicrobial resistance ABSTRACT Curcumin, a natural polyphenolic pigment, has gained significant attention due to its potent antimicrobial properties. However, its applications are limited by poor solubility and bioavailability. Conjugation with an appropriate nanocarrier can enhancing curcumin’s therapeutic efficacy. Carboxylated graphene quantum dots (cGQDs) were chosen as the nanocarrier. cGQDs were functionalized with polyethylene glycol (PEG) and curcumin, resulting in the cGQDs-PEG-curcumin nanocomposite. The antimicrobial photodynamic therapy (aPDT) efficacy of the cGQDs-PEG-curcumin was evaluated against both Gram-positive and Gram-negative bacterial strains, including resistant variants. The cGQDs-PEG-curcumin nanocomposite demonstrated significant antimicrobial efficacy under light activation, particularly against Gram-positive bacterial strains, including resistant ones (Staphylococcus aureus MIC 3.5 µM, MBC 15 µM, Enterococcus faecalis MIC 0.94 µM, MBC 3.75 µM). It also has an effect on Gram-negative bacterial strains (Escherichia coli MIC 30 µM, MBC -). Among resistant strains, the highest efficacy was observed against vancomycin-resistant Enterococcus faecium (MIC 3.75 µM and MBC 3.75 µM) and rez. P. aeruginosa (MIC 15 µM and MBC 30 µM) upon irradiation at 5 J/cm 2 . When double irradiation (2 ×5 J/cm 2 ) was applied, both MIC and MBC values decreased for all tested Gram-positive bacterial strains. Binding curcumin to GQDs overcame the biological barrier posed by its poor water solubility. The synthesized nanocomposite accumulated in the bacterial cell wall, and upon light activation, induced rapid bacterial cell death immediately after irradiation, remarkably without generating reactive oxygen species (ROS). Overall, the cGQDs-PEG-curcumin nanocomposite, as a novel third-generation photosensitizer, demonstrates significant potential for aPDT, with a promising ability to overcome bacterial resistance. 1. Introduction The increasing ability of bacteria to develop resistance to antibiotics poses a serious threat to modern medicine [1]. Even conventional treatments are becoming less effective. If this trend continues, infections caused by multidrug-resistant bacteria may become the leading cause of death in the coming decades (www.who.int), [2]. The global rise in antibiotic resistance among various strains of Gram-positive and Gram-negative bacteria underscores the urgent need for innovative antimicrobial therapies. Besides developing new antibiotics, researchers are developing novel approaches and combinations of existing drugs and treatment methodologies [2,3]. One such promising strategy is antimicrobial photodynamic therapy (aPDT). This method has gained attention due to its minimal invasiveness, effective tissue penetration, reduced damage to healthy tissue, lower systemic toxicity, and potential to overcome chemoresistance [4]. Although photodynamic therapy (PDT) is traditionally used against malignant and benign tumors, many photosensitizers (PSs) also * Corresponding authors. E-mail addresses: [email protected] (K. Bartoˇ n Tom´ ankov´ a), [email protected] (H. Dilenko). Contents lists available at ScienceDirect Biomedicine & Pharmacotherapy journal homepage: www.elsevier.com/locate/biopha https://doi.org/10.1016/j.biopha.2025.118696 Received 31 July 2025; Received in revised form 21 October 2025; Accepted 21 October 2025 Biomedicine & Pharmacotherapy 193 (2025) 118696 0753-3322/© 2025 The Author(s). Published by Elsevier Masson SAS. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). demonstrate potent antimicrobial properties, showing effectiveness against microbial infections, including bacteria, fungi, and viruses [5–8]. Recent studies have revealed that aPDT not only offers local antimicrobial and anti-inflammatory effects, but also modulates immune responses, reducing inflammation and accelerating healing. It can even influence nociception by releasing neurotransmitters and inhibiting nerve impulse conduction, thereby relieving pain [9]. Unlike antibiotics that typically target specific cellular components or metabolic pathways, aPDT primarily works through the generation of ROS, energy transfer, and localized oxidative damage to various cellular structures. This non-specific mechanism significantly reduces the likelihood of microbial resistance development [10]. Moreover, aPDT enables precise targeted treatment of infected areas via light activation, minimizing harm to surrounding healthy tissues and reducing systemic side effects. The therapy can also be combined with antibiotics or antifungal agents to enhance therapeutic outcomes and potentially shorten treatment duration [11,12]. Furthermore, aPDT is considered environmentally benign, as it relies on light and photosensitizers that do not contribute to chemoresistance or environmental contamination, unlike traditional antimicrobial agents [13]. The principle of PDT involves activating a photosensitizer with light of a specific wavelength, leading to the generation of ROS with strong antimicrobial activity [14]. This mechanism typically relies on oxygen-dependent pathways, classified as type I or type II PDT. In these pathways, ROS inflict oxidative damage that ultimately destroys microbial cells. An oxygen-independent pathway, sometimes referred to as type III PDT, has also been described. In this case, a photosensitizer in its excited state directly interacts with biomolecules such as proteins and nucleic acids, causing their degradation [15]. Despite the remarkable antibacterial properties of many sensitizers, their clinical application is often hindered by unfavorable physicochemical characteristics, such as poor solubility, low photodynamic stability, suboptimal biodistribution, and limited biocompatibility [16–18]. Curcumin, a natural photosensitizer, also shares these properties but remains a molecule of significant interest due to its wide range of therapeutic activities. These include anti-inflammatory, antioxidant, chemopreventive, chemosensitizing, radiosensitizing, antimicrobial, anticancer, and wound-healing effects [19–22]. Curcumin is structurally a hydrophobic diferuloylmethane derivative compound featuring two aromatic rings linked by a conjugated carbon chain. It contains two α ,β-unsaturated carbonyl groups essential for its biological activity. Despite its hydrophobicity, the presence of phenolic and carbonyl groups enables hydrogen bonding interactions. Studies have demonstrated that curcumin interacts with a variety of biomolecules, including signaling molecules, protein kinases, protein reductases, HIV-1 integrase and protease, DNA, and RNA [23–27]. Graphene quantum dots (GQDs) are emerging as promising drug carriers due to their unique properties [28,29]. GQDs are nanoscale fragments of graphene, typically less than 100 nm in size (0-dimensional graphene material), consisting of one to several layers of graphene, and can be modified with various functional groups [23]. They have unique optical and chemical properties such as large surface area, tunable size, good photostability, thermal properties, biocompatibility, low cytotoxicity, facilitate targeted drug delivery, and improve therapeutic outcomes [30]. GQDs can be used in applications such as bioimaging, biosensing, and notably in photodynamic therapy (PDT) or photothermal therapy (PTT) [31–34]. These properties make GQDs ideal candidates for developing advanced therapeutic drugs. Among the various surface modifications explored to optimize the functionality of GQDs, the introduction of carboxyl groups has been particularly effective in enhancing their solubility, biocompatibility, and versatility for biomedical applications. Carboxylated GQDs (cGQDs) can be used as carriers for drugs and biomolecules, sensors for detecting a range of analytes, catalysts for chemical reactions, and more [35–37]. Their unique properties continue to drive intensive research, with new applications and opportunities steadily emerging. Although GQDs possess inherent photodynamic and photothermal properties, their combination with a photodynamically active compound significantly enhances their effectiveness in photodynamic applications [16,38,39]. cGQDs in combination with a photosensitizer (PS) can effectively combat resistant infections by improving the delivery, stability, and photodynamic activity and overall effectivity of PSs. In this study, we explored the use of cGQDs as a nanocarrier for curcumin, a water-insoluble compound with potent antimicrobial and photodynamic activity. The combination of curcumin with cGQDs led to the development of an advanced nano-delivery system for antimicrobial photodynamic therapy (aPDT), offering several advantages, including improved solubility and stability of curcumin, protection from rapid degradation, enhanced bioavailability, and thereby increased therapeutic efficacy. To date, no studies have reported the use of carboxylated graphene quantum dots (cGQDs) as nanocarriers for curcumin in the context of antimicrobial photodynamic therapy. While curcumin and graphene-based nanomaterials have each been explored separately, their integration into a PEG-functionalized cGQDs-curcumin nanocomposite introduces a novel third-generation photosensitizer with unique properties. This platform not only enhances the solubility, stability, and bioavailability of curcumin, but also demonstrates potent antibacterial efficacy even against multidrug-resistant strains. Importantly, our results reveal an unconventional, ROS-independent mechanism of bacterial inactivation, which sets this system apart from traditional PDT approaches that rely on singlet oxygen production. By highlighting these distinct features, our study provides a new and original contribution to the development of advanced nanomaterials for combating antimicrobial resistance. 2. Material and methods 2.1. Preparation of cGQDs-PEG-curcumin nanocomposite The cGQDs-PEG-curcumin nanocomposite was prepared via a sequential fabrication method. First, polyethylene glycol (PEG 300, Merc) was added to carboxylated graphene quantum dots (cGQDs, ASC Material, USA) to form a 5 % solution, which was stirred for 24 h. Next, an excess amount of curcumin (2 mg, Merc) was introduced to the mixture and stirred for an additional 24 h. To remove unbound curcumin, the mixture was centrifuged at 13,500 rpm for five minutes, and the resulting supernatant was collected for subsequent analyses. 2.2. Characterization of cGQDs-PEG-curcumin nanocomposite 2.2.1. UV-Vis spectrometry UV-Vis spectrometry was used to determine key characteristics of curcumin in the composite, including its absorption maximum and concentration, based on a calibration curve prepared from curcumin dissolved in DMSO. Measurements were performed using a Tecan Infinite 200 Pro microplate reader to ensure precise absorbance readings across a range of concentrations. DMSO was used as a solvent due to curcumin’s poor solubility in aqueous solutions. 2.2.2. Fourier transform infrared spectroscopy (FTIR) The FTIR spectra were recorded using a Nicolet iS5 FTIR spectrometer with a ZnSe crystal and the iD Foundation adapte. The FTIR spectra were recorded and analyzed using OMNIC software, where baseline correction was carried out. The sample was deposited in liquid form onto the FTIR crystal and allowed to dry prior to measurement. 2.2.3. X-ray photoelectron microscopy (XPS) The XPS measurements were carried out using a Thermo Scientific Nexsa G2 surface analysis system equipped with monochromatic Al K α X-ray source (h ν =1486.6 eV). The base pressure in the analysis chamber was maintained below 1 ×10 −7 mbar during all K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 2 measurements. Survey spectra were recorded at a pass energy of 100 eV to identify elemental composition, while high resolution spectra were collected at a pass energy of 20 eV with an energy step size of 0.05 eV. The binding energy scale was calibrated with respect to the adventitious carbon C 1 s peak at 284.8 eV. To compensate for surface charging, a low energy ion flood gun was employed. Data processing and peak fitting were performed using Thermo Avantage software. Background subtraction was achieved using the Smart method, and all peaks were modeled with a mixed Gaussian-Lorentzian line shape. 2.2.4. Fluorescence spectroscopy Fluorescence spectroscopy was conducted using an FLS980 fluorescence spectrometer (Edinburgh Instruments, UK) featuring an R928P photomultiplier tube (Hamamatsu, Japan) housed in a thermoelectrically cooled unit. A 450 W xenon arc lamp served as the excitation source for steady-state measurements. Time-resolved fluorescence was measured using an EPL-375 picosecond pulsed diode laser (λ em = 372 nm; Edinburgh Instruments, UK), coupled with a time-correlated single-photon counting (TCSPC) system. 2.2.5. Transmission electron microscopy (TEM) The ultrafine structure of the synthesized nanomaterials was studied using a TITAN G2 transmission electron microscope (ThermoFisher) operated at an accelerating voltage of 300 kV. A dilute dispersion of the sample (~0.1 mg/ml) prepared by sonication was drop-cast onto a gold grid with a holey carbon film and allowed to dry at room temperature for 24 h prior to analysis. 2.3. Antibacterial study 2.3.1. Antibacterial photodynamic therapy (aPDT) The antibacterial efficacy of aPDT was evaluated against both standard and antibiotic-resistant bacterial strains. The Gram-positive strains Staphylococcus aureus CCM 4223 (ATCC 29213) and Enterococcus faecalis CCM 4224 (ATCC 29212), along with the Gram-negative strain Escherichia coli CCM 3954 (ATCC 25922), were obtained from the Faculty of Science, Masaryk University (Brno, Czech Republic). In addition, antibiotic-resistant strains were included in the study: Vancomycinresistant Gram-positive strain Enterococcus faecium VRE 419/ANA (VanA) and multidrug-resistant Gram-negative strain Pseudomonas aeruginosa 21425, both provided by the culture collection of the Department of Microbiology, Faculty of Medicine and Dentistry, Palacký University Olomouc (Olomouc, Czech Republic). Bacterial strains were cultured on Mueller-Hinton agar for 24 h and were then resuspended in 2 ml of physiological saline to reach an optical density corresponding to the McFarland standard of 1. After further dilution, 1 µl of this suspension was added to each well of the microtiter plate containing 100 µl MH medium, yielding a final inoculum concentration of 5 ×10 5 CFU/ml. The samples were divided into three test groups: 1) Light control (light toxicity): bacterial samples were irradiated with 414 nm light at doses up to 5 J/cm 2 in the absence of a nanocomposite to evaluate lightinduced effects. 2) Dark control (dark toxicity): bacterial samples were incubated with the nanocomposite (0–30 µM) without light exposure to assess potential toxicity in the absence of irradiation. 3) Photodynamic treatment (aPDT group): bacterial samples were incubated with nanocomposite cGQDs-PEG-curcumin (0–30 µM) and cGQDs-PEG, and irradiated with 414 nm light. Irradiation regimes included either 1 ×5 J/ cm 2 , 2 ×5 J/cm 2 , or a single dose of 30 J/cm² (the latter used only for MIC and MBC determination). To assess the uptake of cGQDs-PEG-curcumin nanocomposite by bacterial cells, the labeled bacterial suspensions were washed three times with PBS and centrifuged at 3000 rpm. Afterward, aPDT was induced, and MIC, MBC, and microscopic analysis were performed. All groups were incubated for two hours in a thermobox, with or without the nanocomposite, prior to light exposure. Irradiation was performed using a custom-built LED device (patented, CZ 302829 B6), equipped with 414 nm LEDs arranged in a hexagonal matrix. Each triplet of neighboring LEDs formed an equilateral triangle to ensure homogeneous illumination. The LEDs were mounted on a board positioned opposite the sample, with distance elements ensuring uniform light distribution across the microtiter plate. 2.3.2. Determination of MIC and MBC The antibacterial activity of the nanocomposite was evaluated by determining minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) using the standard microdilution method, following EUCAST guidelines (The European Committee on Antimicrobial Susceptibility Testing – EUCAST, http://www.eucast. org). MIC is defined as the lowest concentration of the antimicrobial agent that inhibits bacterial growth, while MBC refers to the lowest concentration that results in microbial death. To perform the assay, cGQDs-PEG-curcumin was serially diluted in Mueller-Hinton (MH) medium within microtiter plates to obtain a final concentration ranging from 0 to 30 µM. Bacterial strains were added to the wells and incubated for two hours. Following incubation, samples were irradiated under defined conditions. After light exposure, plates were further incubated for 18 h at 35 ±1 ◦C. MIC values were recorded as the lowest concentrations at which no visible bacterial growth was observed. To determine MBC values, 10 μ l aliquots from wells without visible growth were subcultured onto MH agar plates and incubated overnight at 35 ±1 ◦C. MBC was defined as the lowest concentration at which no bacterial colonies were observed on the agar surface. 2.3.3. Evaluation of bactericidal effect in time (Time-Kill Assay) The time-dependent bactericidal activity of cGQDs-PEG-curcumin was evaluated using a time-kill assay conducted in microtiter plates. The nanocomposite was diluted in MH medium to final concentrations corresponding to 2 ×MIC, MIC, ½ MIC and 0 mg/L (control). The plates were inoculated with strain Staphylococcus aureus CCM 4223 at an initial concentration of 10 5 CFU/ml. Following a two-hour incubation at 35 ±1◦C, samples were irradiated with 414 nm light at a dose of 5 J/cm 2 to initiate aPDT. Bacterial activity was assessed at multiple time points: two hours before irradiation (baseline), immediately after irradiation (0 min), and then at 15, 30, 60, 120 min, and 18 h post-aPDT. At each time point, aliquots were withdrawn, serially diluted in physiological saline and inoculated onto MH agar. After incubation for 18 ±2 h at 35 ±1 ◦C, colony-forming units (CFUs) were counted. Results were expressed as CFU/ml, and time-kill curves were constructed. The limit of detection (LOD) for the assay was log 1 CFU/ml [40]. 2.3.4. Determination of bacterial survival – growth curve Bacterial survival was evaluated by monitoring growth curves following treatment. In all cases, 100 µl of bacterial suspension (5 ×10 5 cells/ml) was incubated in the dark at 37◦C for two hours with cGQDsPEG-curcumin (0–30 µM). After the dark incubation, photodynamic therapy (PDT) was applied, and absorbance was measured immediately after irradiation and then hourly for 24 h using a BioTek spectrophotometer (USA) equipped with Gen5 software. Measurements were taken at a wavelength of 630 nm. The photoinactivation efficacy of cGQDsPEG-curcumin against bacterial strains was assessed by comparing the optical density (OD) of treated samples to that of control groups (untreated cells). Changes in OD over the 24-hour period provided a quantitative measure of bacterial survival and growth inhibition following aPDT. 2.3.5. ROS and singlet oxygen detection ROS and singlet oxygen generation were assessed using the fluorescence probes CM-H 2 DCFDA and SOSG, respectively. Briefly, a bacterial suspension (5 ×10 6 cells/ml) in PBS was cultured with the cGQDs-PEGcurcumin nanocomposite (0–30 µM) and either 5 µM CM-H 2 DCFDA or SOSG in a 96-well plate at 35 ±1 ◦C for two hours. After incubation, the K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 3 samples were irradiated with monochromatic light at 414 nm (5 J/cm 2 ). ROS and singlet oxygen generation were measured immediately before and post-irradiation using a Synergy HT microplate reader by recording CM-H 2 DCFDA fluorescence at excitation/emission wavelengths of 493/ 520 nm, and SOSG fluorescence at 504/525 nm. The results were expressed as percentages relative to the untreated control group. 2.3.6. Determination of lipid peroxidation Lipid peroxidation was assessed using the fluorescent probe BODIPY. A bacterial suspension (5 ×10 6 cells/ml) in PBS was cultured with cGQDs-PEG-curcumin nanocomposite (0–30 µM) and 1 µM BODIPY in a 96-well plate at 35 ±1 ◦C for two hours. Following incubation, the samples were irradiated with monochromatic light at 414 nm (5 J/cm 2 ). Fluorescence measurements were taken one and four hours postirradiation using a Synergy HT microplate reader. The oxidation state of lipid structures was indicated by the green fluorescence signal (excitation/emission: 488/510 nm), while the reduction state was indicated by the red fluorescence signal (excitation/emission: 581/ 591 nm). The ratio of red to green fluorescence was used to quantify the extent of lipid peroxidation, with lower red/green ratios indicating higher levels of peroxidation. Data were expressed as a percentage relative to the untreated control group. 2.3.7. Live-dead determination – bacterial wall integrity The bacterial cell suspension (5 ×10 6 cells/ml) was cultured in a 96well plate in MH medium, at 35 ±1 ◦C for 24 h. Then the cGQDs-PEGcurcumin was added to the bacteria at a concentration of 0–30 µM. After 2 h incubation, the photodynamic effect was induced using a 414 nm irradiator in a light dose of 5 J/cm 2 . Changes of bacterial wall integrity were measured 1 and 4 h after irradiation using Live/dead BacLight bacterial viability kit (Invitrogen). 25 µl of staining mixture (6 µl SYTO9, 6 µl PI, 2 ml H 2 O) was added to 25 µl of bacteria strain in a 96-well plate with ultra-thin bottom (Ibidi) and incubated for 15 min in the dark. Ex/ Em for SYTO9 was set to 485/530 nm and for PI 485/630 using the microplate reader Synergy HT. The ratio between green and red fluorescence indicates the extent of bacterial wall damage. The data were expressed as a percentage relative to the control group. 2.4. Microscopy study 2.4.1. HR confocal microscopy – localization of cGQDs-PEG-curcumin in microbial Body High-resolution confocal microscopy (LSM 780, Zeiss) equipped with ZEN software was used to investigate the localizations of cGQDs-PEGcurcumin in the microbial body. Bacterial suspension with a concentration of 10 6 bacteria per well was incubated with 15 µM of the nanocomposite in 96-well plates for 24 h at 35 ±1 ◦C. To refine the visualization of the localization of the composite, samples of bacterial suspension were stained with Rhodamine B (RhoB, RBITC) at a final concentration of 1 µM for 20 min. To remove unbound dye, the suspensions were centrifuged twice at 3000 rpm for three3 minutes and washed with PBS. Two staining protocols were used to identify the colocalization: 1) incubation with the cGQDs-PEG-curcumin nanocomposite for 24 h prior to the addition of RhoB; 2) pre-staining with RhoB followed by the addition of the cGQDs-PEG-curcumin nanocomposite. For immobilization, 10 µl of bacteria suspension was mixed with 25 µl of low-melting-point (LMP) agarose in PBS and transferred onto a microscope slide, then covered with a cover glass. Fluorescence measurements were performed at 405 nm excitation for cGQDs-PEGcurcumin and 560 nm excitation for RhoB. 2.4.2. Scanning electron microscopy (SEM) To evaluate bacterial morphological changes induced by cGQDsPEG-curcumin nanocomposite, scanning electron microscopy was performed using Scios 2 (ThermoFisher Scientific). Bacterial suspensions (10 6 CFU/well) were incubated with the nanocomposite at concentrations of 10, 20, and 40 µM for two hours at 35 ±1 ◦C in 96-well plates. Each condition, including untreated controls, was prepared in quintuplicate. Two identical plates were prepared: one to evaluate dark toxicity and the other for PDT. After incubation, samples from the PDT group were irradiated with 414 nm light at a dose of 5 J/cm 2 . Bacterial suspensions were then transferred to Eppendorf tubes, centrifuged (3000 RPM, 5 min), and washed four times with 1 ml of distilled water. The resulting pellet was resuspended in 50 µL of distilled water, and 10 µL of the suspension was applied to a pre-disinfected aluminum SEM pin. Following drying, the pins were heat-fixed by briefly passing them three times through the non-luminous part of a flame to enhance bacterial adhesion to the surface of the pin. 2.4.3. Statistical analysis All experiments were performed in triplicate. Graphs and statistical analyses were conducted using GraphPad Prism software. Two-way analysis of variance (ANOVA) was employed to evaluate differences between groups, followed by Tukey’s post hoc analysis (a multiple comparisons test), to assess the significance of the data. 3. Results 3.1. Preparation and characterization of cGQDs-PEG-curcumin nanocomposite 3.1.1. Synthesis of cGQDs-PEG-Curcumin Nanocomposite The cGQDs-PEG-curcumin nanocomposite was successfully synthesized via a simple two-step method involving PEGylation of carboxylated graphene quantum dots followed by curcumin loading. The use of PEG not only enhances the solubility and biocompatibility of the composite, but also likely facilitates curcumin attachment through noncovalent interactions, such as hydrogen bonding and π – π stacking. The excess of curcumin ensured maximal loading, while the final centrifugation step effectively removed unbound drug, allowing for the collection of a stable and homogeneous nanocomposite. This straightforward and reproducible approach demonstrates potential for scalable production of photodynamically active nanomaterials for biomedical applications. 3.1.2. Determination of curcumin concentration The absorption maximum of curcumin conjugated to the cGQDs and functionalized with PEG exhibited a blue shift to 416 nm, compared to 436 nm for pure curcumin dissolved in DMSO. This spectral shift is likely attributed to interactions between curcumin and the cGQDs, resulting in alteration to the electronic environment of curcumin upon conjugation. Additionally, the observed shift may be partially influenced by the solvent effect, as pure curcumin is poorly soluble in water and its absorbance was measured in DMSO [41]. 3.1.3. Structural and morphological characterization, fluorescence spectroscopy The structural features of the cGQDs-PEG-curcumin nanocomposite were analyzed by FTIR spectroscopy and compared with those of its individual components. The spectrum displayed characteristic absorption bands of curcumin, including a broad peak at 3200–3500 cm⁻¹ corresponding to hydroxyl groups, C–H stretching near 2900 cm⁻¹ , carbonyl (C – – O) vibrations in the 1600–1650 cm⁻¹ region, and aromatic C – – C stretching between 1450–1600 cm⁻¹. These findings confirm the successful conjugation of curcumin onto the PEGylated cGQDs surface [41]. X-ray photoelectron spectroscopy (XPS) was performed to analyze the elemental composition and surface chemical states of cGQDs-PEGcurcumin nanocomposite and its constituents. The XPS survey spectrum of cGQDs (Fig. 1B) revealed the presence of two dominant elements, carbon (C 1 s) and oxygen (O 1 s), confirming their composition K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 4 as carbon-based nanomaterials with oxygen-containing surface functionalities. The high-resolution C 1 s spectrum was deconvoluted into three main components centered at approximately 284.8 eV (C–C), 286.2 eV (C–O), and 289.0 eV (O–C – – O), corresponding to graphitic, hydroxyl, and carboxyl groups, respectively. The relative intensity of oxygenated carbon species indicates a high degree of surface oxidation, which contributes to the hydrophilicity and chemical reactivity of cGQDs. These oxygen-containing groups provide active sites for subsequent PEGylation and curcumin conjugation. The XPS survey spectrum of curcumin (Fig. 1 C) exhibited characteristic signals of carbon (C 1 s) and oxygen (O 1 s), consistent with its organic molecular structure. The high-resolution C 1 s spectrum was deconvoluted into peaks at approximately 283.5 eV (C–H) corresponding to aliphatic C–H bonds present in the methylene and methyl groups of the curcumin molecule, 284.8 eV (C–C/C – – C) corresponding to aromatic and aliphatic carbons, 286.4 eV (C–N/C–O), and 287.9 eV (C – – N/C – – O) attributed to carbonyl functionalities of the β-diketone moiety. The relatively high oxygen content reflects the presence of hydroxyl and carbonyl groups, which are responsible for curcumin’s polarity and potential for hydrogen bonding. These oxygenated functional groups facilitate subsequent interactions Fig. 1. (A) FTIR spectra of cGQDs, PEG, curcumin and cGQDs-PEG-curcumin nanocomposite. XPS spectra of (B) cGQDs, (C) curcumin and (D) cGQDs-PEG-curcumin nanocomposite, with insets showing high-resolution C 1 s spectra (left) and elemental compositions (right). Excitation-emission colour maps of (E) cGQDs, (F) curcumin, (G) PEG, and (H) cGQDs-PEG-curc0075min nanocomposite. PL decay curves of (I) cGQDs (λ em =455 nm) and (J) curcumin (λ em =545 nm) in neat form and in the composite. HRTEM images of (K) cGQDs, (L) curcumin in DMSO, and (M) cGQDs-PEG-curcumin nanocomposite. K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 5 with PEG and cGQDs during nanocomposite formation. The XPS survey spectrum of the cGQDs-PEG-curcumin nanocomposite (Fig. 1D) showed predominant C 1 s and O 1 s signals, confirming the carbon-oxygen-rich surface composition. Compared to pristine cGQDs, the high-resolution C 1 s spectrum exhibited notable changes in peak distribution, reflecting successful surface functionalization. The characteristic carboxyl (O–C – – O) peak observed in cGQDs at ~289.0 eV was no longer detected, indicating that the surface carboxyl groups were consumed during covalent coupling with PEG and curcumin. New component appeared at ~287.0 eV (C – – O), corresponding to carbonyl bonds derived from PEG chains and curcumin moieties. These spectral modifications confirm the effective conjugation of curcumin onto the PEGylated cGQD surface through the replacement of free carboxyl groups by amide linkages. Excitation-emission color maps of cGQDs (Fig. 1a) and curcumin (Fig. 1b) display distinct spectral patterns: cGQDs exhibit emission centered around 455 nm, while curcumin shows emission near 545 nm. In the hybrid nanocomposite cGQDs-PEG-curcumin sample (Fig. 1d), both emission features are clearly present, confirming the successful incorporation of both components into the composite. The excitationemission map of pure PEG (Fig. 1c) shows fluorescence around 310 nm. However, this signal is absent in the nanocomposite sample, indicating that PEG contributes negligibly to overall fluorescence due to its low concentration in the hybrid structure. PL decay measurements at the emission wavelength of 455 nm (corresponding to cGQDs) revealed a pronounced decrease in PL lifetime upon formation of the hybrid nanocomposite. For the cGQDs-PEGcurcumin sample, the dominant PL lifetime component was τ 1 =0.32 ns with an amplitude of B 1 =73 %, compared to τ 1 =0.73 ns and B 1 =39 % in the neat cGQDs sample (Fig. 1e). This reduction in PL lifetime suggests a strong interaction between cGQDs and the overcoated PEG-curcumin layer, most likely due to charge or energy transfer processes occurring at the interface. Additionally, the amplitude of the dominant PL lifetime component of curcumin in the nanocomposite decreased to 88 %, compared to 94 % in the neat curcumin sample, further supporting the occurrence of intermolecular interactions within the composite structure. As shown in Fig. 1 K, the commercially obtained cQGDs exhibited particle sizes ranging from 10 to 15 nm. Curcumin, which is purely soluble in polar solvents but readily dissolves in organic solvents, was dissolved in DMSO for morphological analysis. After drying, the curcumin structure was imaged and is shown in Fig. 1 L. Following PEGylation of cQGDs and subsequent conjugation with curcumin, the resulting hybrid nanocomposite exhibited a marked reduction I particle size, with average diameters around 3 nm (Fig. 1 M). This size reduction is most likely attributable to structural reorganization during PEGylation and the sonication process used during preparation of the nanocomposite [41]. 3.2. Antibacterial mechanism of aPDT 3.2.1. Determination of MIC and MBC cGQDs-PEG-curcumin exhibited a significant antibacterial photodynamic effect under a light dose of 5 J/cm 2 against both Gram-positive and Gram-negative bacterial strains, including resistant strains. For Gram-positive bacteria, the following MIC and MBC were observed: S. aureus (MIC: 3.75 µM, MBC: 15 µM), E. faecalis (MIC: 0.94 µM, MBC: 3.75 µM), and E. faecium VRE (MIC: 3.75 µM, MBC: 3.75 µM). For Gramnegative bacteria, E. coli showed the MIC of 30 µM, but no MBC could be determined, while P. aeruginosa exhibited MIC and MBC values of 15 µM and 30 µM, respectively (see Table 1). The application of a second light dose two hours after the initial irradiation led to enhanced antibacterial effects in Gram-positive strains. Specifically, for S. aureus, the MIC decreased to 1.85 µM and the MBC to 3.75 µM. For both E. faecalis and E. faecium VRE, the MIC and MBC values were reduced to 0.94 µM. A single higher dose of 30 J/cm² was also evaluated; however, the MIC and MBC values remained similar to those observed with 5 J/cm², indicating no additional improvement in efficacy. Interestingly, both resistant strains, E. faecium VRE and P. aeruginosa, were also sensitive to the nanocomposite in the absence of light, exhibiting a MIC of 30 µM due to dark toxicity. These results confirm that cGQDs-PEG-curcumin exerts a strong antibacterial effect, particularly against Gram-positive bacteria. The antibacterial activity of the cGQDs-PEG was also investigated. MIC and MBC values were determined, and the results demonstrated that the composite lacking curcumin exhibited no inhibitory effect against any of the bacterial strains examined. To investigate the mechanism of action, bacterial suspensions labeled with cGQDs-PEG-curcumin were washed three times with PBS and centrifuged prior to aPDT to remove unbound nanocomposite. No photodynamic antibacterial effect was observed after this washing step, and MIC and MBC values could not be determined. Bacterial growth corresponded to the control group, suggesting that the presence of free nanocomposite is essential for the aPDT effect. 3.2.2. Evaluation of bactericidal effect over time (Time-Kill Assay) The results are shown in Fig. 2. The bactericidal effect of cGQDsPEG-curcumin against S. aureus was shown to be dependent on both the concentration of the nanocomposite and light irradiation. At concentrations equal to or exceeding the MIC, bacterial growth was immediately inhibited during irradiation, demonstrating a rapid bactericidal response. At sub-MIC concentrations, irradiation significantly reduced the number of viable bacteria (by approximately two orders of magnitude). However, this reduction was not sustained. After 18 h of incubation, the bacterial count returned to levels comparable to the untreated growth control. 3.2.3. Determination of bacterial survival – growth curve Bacterial survival was assessed by monitoring absorbance at 630 nm over 24 h following photodynamic treatment with cGQDs-PEGcurcumin (0–30 µM) and a light dose of 5 J/cm². The experiment was Table 1 MIC and MBC (µM) values for selected bacterial strains following aPDT with cGQDs-PEG-curcumin nanocomposite. 0 J/cm 2 5 J/cm 2 2×5 J/cm 2 cGQDs-PEG-curcumin (µM) MIC MBC MIC MBC MIC MBC Staphylococcus aureus - - 3.75 15 1.87 3.75 Enterococcus faecalis - - 0.94 3.75 0.94 0.94 Enterococcus faecium VRE 30 - 3.75 3.75 0.94 0.94 Escherichia coli - - 30 - 30 - rez. Pseudomonas aeruginosa 30 - 15 30 15 30 Fig. 2. Bactericidal effect over time (time-kill assay) of cGQDs-PEG-curcumin after aPDT under a light dose of 5 J/cm 2 . K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 6 conducted on S. aureus and E. coli strains. The growth curves (Fig. 3) reflect distinct phases, allowing for the interpretation of bacterial viability post-treatment. In all samples, an initial lag phase was observed, during which no significant change in absorbance occurred. This was followed by an exponential growth phase, especially prominent in control samples and in those treated with lower concentrations of cGQDs-PEG-curcumin. Reduced absorbance in treated samples indicates a decrease in viable bacterial cells due to the effect of aPDT. Control groups (without photosensitizer or irradiation) showed a continuous increase in absorbance, corresponding to typical bacterial proliferation (Fig. 3). A more pronounced reduction in absorbance over time was observed in S. aureus (Fig. 3A). A sustained drop in absorbance was detected approximately 9 h after aPDT in samples treated with higher concentrations (7.5–30 µM), indicating effective inhibition of bacterial growth. Absorbance remained low for up to 24 h, confirming reduced viability. At a concentration of 3.75 µM, delayed bacterial growth was observed, beginning approximately 21 h after treatment. In contrast, lower concentrations (0.94–1.87 µM) showed regrowth starting as early as 9 h after treatment. Partial survival was still evident at 24 h, which is consistent with the higher MBC values obtained after 48 h of incubation. Fig. 3B illustrates the response of the Gram-negative E. coli strain under the same treatment conditions. A sustained reduction in viability was observed at 15 and 30 µM, though regrowth occurred at all concentrations except for 30 µM. Recovery typically began 10–15 h after treatment. As a result, the MBC for E. coli could not be determined. 3.2.4. ROS and singlet oxygen production The levels of ROS following aPDT treatment are illustrated in Fig. 4. All data are normalized to 100 % of the untreated control. The results indicate that aPDT with cGQDs-PEG-curcumin does not lead to an increase in ROS levels in all the tested bacterial strains when compared to the negative control (C-). The only exception was observed in E. faecium VRE, where ROS production occurred under 414 nm light irradiation in the positive control group (C+). Interestingly, a reduction in ROS/singlet oxygen levels was observed in all bacterial strains following treatment, with the effect being more pronounced after aPDT induction. This suggests that cGQDs-PEGcurcumin exhibits antioxidant behavior. At concentrations of 15 and 30 µM, ROS levels decrease to as low as 67 % relative to the light and nanocomposite negative control. The SOSG measurement shows the same trend of 1 O 2 production as in the ROS assay. The overall increase in 1 O 2 levels in the 5 J/cm 2 group including the control group compared to the non-irradiated group (Fig. 4 C, D) is due to the sensitivity of this fluorescent probe to the light. Notably, the 30 μ M concentration in the non-irradiated condition elicited a 20 % elevation in singlet oxygen (¹O₂) levels in E. aureus strains and a 16 % elevation in E. faecalis. However, this increase does not correspond with the MIC and MBC results and, over the long term, exerts no measurable effect on the viability of the respective strains. These findings indicate that the nanocomposite displays antioxidant properties in both its non-irradiated and irradiated forms, potentially modulating oxidative stress in bacterial environments rather than enhancing it through ROS generation. 3.2.5. Determination of lipid peroxidation Lipid peroxidation was assessed using the fluorescent probe BODIPY, as shown in Fig. 5. The results are presented as the ratio of red to green fluorescence (R/G), which reflects the balance between the reduced (non-oxidized) and oxidized forms of lipids. A lower R/G ratio indicates higher levels of oxidative damage to membrane lipids. Measurements were taken both with and without light irradiation. The results demonstrate that the R/G ratio increased with rising concentrations of cGQDs-PEG-curcumin, indicating reduced lipid peroxidation and aligning with the decreased ROS levels observed in Fig. 4. The lowest levels of oxidative damage were detected at the highest concentrations tested (15 and 30 µM). Following aPDT (hatched bars), lipid peroxidation levels were frequently lower than in untreated controls, as evidenced by a higher R/G ratio. This suggests a protective antioxidant effect of cGQDs-PEG-curcumin, which may play a key role in stabilizing bacterial lipid membranes against oxidative stress, especially at higher concentrations. Importantly, in none of the tested conditions did the R/G fluorescence ratio fall below 100 % of the non-irradiated control (C-), further supporting the notion that the composite does not enhance lipid oxidative damage under the tested conditions. 3.2.6. Live-dead determination – assessment of bacterial wall integrity Fluorescent dyes such as propidium iodide (PI) and SYTOX are widely used to evaluate bacterial membrane integrity. These dyes are impermeable to intact bacterial membranes, but can penetrate damaged membranes and bind to nucleic acids. The fluorescence signal indicates membrane damage. Green fluorescent dye SYTO 9 is used to stain live bacteria with intact membranes, whereas propidium iodide is used to stain bacteria with damaged membranes. Lower green/red (SYTO 9/PI) fluorescence ratio indicates greater membrane damage. Following aPDT with cGQDs-PEG-curcumin, a significant reduction in membrane integrity was observed in all tested bacterial strains compared to the control groups. As shown in Fig. 6, the most pronounced damage occurred in E. faecium VRE, followed by S. aureus (with a ratio value dropping below 20 % of the control level), and closely followed by E. faecalis. In E. faecium VRE, membrane damage was evident immediately after aPDT induction, while no significant effect was observed without irradiation. Gram-negative strains also showed reduced membrane integrity, though to a lesser extent (60–80 % of control values). This milder effect is likely due to their lower inherent sensitivity to aPDT and the protective structure of the Gram-negative cell wall. The data also suggest that membrane damage increases over Fig. 3. Growth curves of bacterial strain after aPDT induction using cGQDs-PEG-curcumin nanocomposite on S. aureus (A) and E. coli (B). K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 7 time following irradiation, with this time-dependent effect particularly pronounced in S. aureus and E. faecium VRE. 3.3. Microscopy study Localization of cGQDs-PEG-curcumin in microbial body was analyzed using High Resolution confocal microscopy LSM 780, with excitation laser 405 nm. The cGQDs-PEG-curcumin was visualized in all studied bacterial strains (Fig. 7). In no studied bacterial strain was the entry into the intracellular space observed. The nano-composite always accumulated on the surface of bacteria, particularly in areas with increased metabolic activity (septum). Evidence of the lack of penetration into the intracellular space was also provided by the MIC/MBC test after washing off the cGQDsPEG-curcumin. After the induction of aPDT in the washed sample, no reduction in bacterial viability was observed. Due to the hydrophobic nature of curcumin, it is highly likely that the composite accumulates in the lipid components of the cell membrane. However, S. aureus, E. faecium VRE, and E. faecalis have a Grampositive cell wall containing a thicker layer of peptidoglycan, which may affect the binding of the composite. The carboxyl groups of quantum dots may interact with charged components of the cell wall, such as teichoic acids or lipoteichoic acids. These structures are specific to Gram-positive bacteria, are present in the cell wall, and may interact with charged particles through electrostatic interactions. E. coli and P. aeruginosa have a Gram-negative cell wall with an outer membrane containing lipopolysaccharides and lipoproteins, which may provide binding sites for the composite’s hydrophobic components. Interactions with exopolysaccharides forming the biofilm in the studied strains are also possible. Carboxyl groups or PEG could facilitate this binding. Given the hydrophobic nature of curcumin and the properties of carboxyl groups on cGQDs, cGQDs-PEG-curcumin can bind both to lipid components in the membrane and to peptidoglycan structures. This corresponds to the fluorescence signals, which are similar across all studied bacterial strains. To further investigate binding specificity, Rhodamine B (RhoB) staining was employed. RhoB is a cationic fluorescent dye that binds to amino groups in proteins and lipid structures, with its localization influenced by bacterial type and cell wall properties. Due to its hydrophobic regions, RhoB primarily associates with membrane lipids. Two staining protocols were used: (1) bacteria were incubated with cGQDs-PEG-curcumin for 24 h, followed by RhoB staining and PBS Fig. 4. ROS and singlet oxygen generated by bacterial strains incubated with cGQDs-PEG-curcumin without (A, C) and after aPDT induction (B, D). Fig. 5. Lipid peroxidation of bacterial strains without (A) and after aPDT induction (B). The values represent the ratio of the red/green signals, where the red signal stains lipids in their reduced form, and oxidative damage causes a shift in emission to the green region of the spectrum. K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 8 washing (Fig. 8); and (1) bacteria were first stained with RhoB and washed, then incubated with cGQDs-PEG-curcumin without further washing (data not shown). The green fluorescence from cGQDs-PEGcurcumin was stronger in the first approach, suggesting that RhoB may compete for the same binding sites. Co-localization was particularly evident at the septum, where both green (cGQDs-PEG-curcumin) and red (RhoB) signals overlapped. Outside the septum, the two signals were mostly distinct (Fig. 6). 3.3.1. Scanning electron microscopy (SEM) SEM analysis was performed on four bacterial strains (S. aureus, E. faecium VRE, E. coli, and P. aeruginosa) to assess morphological changes following treatment (Figs. 9–16). Two experimental conditions were tested. In the first (dark toxicity), bacteria were exposed to cGQDs-PEGcurcumin at concentrations of 0, 5, 10, and 20 µM for S. aureus and E. faecium VRE, or 0, 20, and 40 µM for E. coli and P. aeruginosa without irradiation. In the second condition, samples were irradiated with a light dose of 5 J/cm² at 414 nm to induce aPDT. In the control groups, bacterial cells appeared morphologically intact. Slight morphological alterations were observed in some nonirradiated samples at higher cGQDs-PEG-curcumin concentrations, particularly in P. aeruginosa. However, pronounced morphological damage was evident after aPDT across all studied strains, including Gram-negative strains. This included disrupted membranes, cell wall rupture, and in some cases, complete fragmentation of bacterial cells into smaller debris. Interestingly, not all cells within a given field were equally affected; some bacteria remained undamaged despite surrounding cellular destruction. In several images, surface regions displaying material evaporation or erosion were visible, likely corresponding to areas where cGQDs-PEG-curcumin had accumulated within the bacterial envelope. These features suggest that the nanocomposite may have been ablated by the electron beam during imaging due to its sensitivity to high-energy exposure. 4. Discussion Curcumin has been widely studied for its therapeutic potential in treating diverse conditions, including HIV, diabetes, osteoporosis, cardiovascular and neurological disorders, and various cancers. It also demonstrates broad-spectrum antiviral and antibacterial activity [42]. However, its clinical application remains limited due to poor pharmacokinetics, low water solubility, and pH-dependent instability. At physiological or alkaline pH (≥7.2), curcumin rapidly degrades, with ~90 % decomposed within 30 min. By contrast, it is more stable in complex biological media such as human serum or cell culture media with 10 % FBS, where less than 20 % of curcumin degrades within 1 h, and approximately 50 % remains after 8 h [43]. Encapsulation or binding of curcumin to nanocarriers has proven effective in improving its stability and solubility [22,44–46]. Our previous study (Dilenko et al., 2025) demonstrated enhanced stability of curcumin when conjugated with carboxylated GQDs. In DMEM, cGQDs-PEG-curcumin remained stable for at least 10 days – five times longer than typical in vitro exposure times. While the hydrodynamic diameter remained 1–5 nm in water, it increased in biological medium to ~20 nm on the first day and to ~48 nm by the third day, indicating aggregation or interaction with amino acids/nutrients in the medium. The zeta potential of the cGQDs-PEG-cur sample in DMEM was around –11 mV, and throughout the 10-day period, it fluctuates slightly, remaining below –10 mV [41]. Free curcumin exhibits potent photodynamic activity, but is prone to photodegradation upon light exposure. Upon light exposure, curcumin degradation products can be detected after just 15 min [43]. Rapid photodegradation has been particularly observed in organic solvents [44]. However, we also assume that curcumin undergoes immediate photoinactivation after being transferred to aqueous solution and exposed to light. Our time-kill assay showed rapid S. aureus inactivation immediately after PDT induction, following two-hour dark incubation with the cGQDs-PEG-curcumin (Fig. 2). The second light dose administered two hours later further enhanced efficacy, lowering the MIC from 3.75 µM to 1.85 µM and the MBC from 15 µM to 3.75 µM (Table 1). A significant effect of double irradiation was observed for E. faecium VRE, with MIC and MBC values dropping from 3.75 µM to 0.94 µM. The other studied Gram-positive strains, however, were not affected by the additional light dose. The minimum bactericidal concentration (MBC) of curcumin in aPDT is influenced by various experimental parameters, including the irradiation intensity and duration, the formulation of curcumin (aqueous or organic), the type of nanocarrier employed, and the bacterial strain targeted. It is generally assumed that increasing either irradiation time at low light intensity or the overall light intensity enhances photodynamic efficacy [47,48]. However, our results did not show any significant difference in MIC or MBC values between light doses of 5 J/cm² and 30 J/cm². Bacterial inactivation by the cGQDs-PEG-curcumin nanocomposite occurred rapidly upon irradiation, and increasing the light did not further improve efficacy. In contrast, administering a second 5 J/cm² light dose led to a notable reduction in both MIC and MBC values, suggesting that repeated light exposure may be more effective than a single high-dose irradiation. The antimicrobial activity of curcumin is strongly dependent on its formulation. In DMSO or ethanol, the average MBC of curcumin for S. aureus without light exposure is approximately [49]. Upon blue light irradiation (605 J/cm²) for 3 h, the MBC drops to ~20 µM [50]. For other strains: E. coli: MBC ~9.6 mM, P. aeruginosa: MBC 13.6–54.3 mM, Fig. 6. Bacterial membrane integrity after aPDT induction with cGQDs-PEGcurcumin nanocomposite at a concentration of 30 µM. Membrane damage was assessed using the SYTO 9/PI staining method. Statistical analysis was performed using a two-way ANOVA, evaluating the effect of bacterial strain (column factor), treatment condition and incubation time (row factor), and their interaction. All three sources of variation were statistically significant: interaction (18.16 % of total variation, ****, p <0.0001), row factor (35.79 %, ****, p <0.0001), and column factor (26.85 %, ****, p <0.0001), indicating a strong influence of both bacterial type and treatment conditions. Tukey’s multiple comparisons test was used as a post hoc analysis to identify specific group differences. Significance is denoted as ** (p <0.01) and **** (p < 0.0001) in the figure. K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 9 platform for curcumin delivery in aPD. cGQDs themselves exhibit photodynamic properties under appropriate light irradiation, and carboxylation has been reported to enhance singlet oxygen ( 1 O 2 ) quantum yield by up to 20-fold [60]. The antimicrobial effectiveness of curcumin or its nanoformulation under aPDT also depends significantly on microbial strain sensitivity. Generally, Gram-positive bacteria are more susceptible, owing to their cell envelope structure, which facilitates better penetration of curcumin compared to Gram-negative bacteria [49]. Gram-positive bacteria possess a thick peptidoglycan layer and an inner plasma membrane but lack an outer protective membrane, allowing easier access to internal targets. Conversely, Gram-negative bacteria have a more complex envelope, comprising an inner membrane, a thin peptidoglycan layer, and an outer membrane enriched in lipopolysaccharides, which serves as a permeability barrier and contributes to osmotic resistance [61]. In Gram-positive strains, the peptidoglycan layer can be directly disrupted, and overall, access to the cytoplasmic membrane is easier. Carboxyl groups on cGQDs or curcumin may interact with teichoic and lipoteichoic acids, weakening the structural integrity of the cell wall and leading to lysis [62,63]. Additionally, interaction with membrane lipids can result in ionic homeostasis [64]. In Gram-negative strains, the carboxyl groups of cGQDs are presumed to interact with phosphate groups or lipids in the lipopolysaccharide layer, compromising the outer membrane. PEGylation may further facilitate the penetration of hydrophobic curcumin into lipid-rich membrane regions, increasing permeability and enabling deeper penetration of the complex into the periplasmic space or cytoplasmic membrane, leading to ion leakage. Curcumin also exhibits chelating properties, binding metal ions such as Fe²⁺, Cu²⁺, and Zn²⁺, thereby disrupting mettalloenzyme function and cellular redox homeostasis, which can result in metabolic collapse [65]. Munir et al. described four principal mechanisms by which curcumin can exert antimicrobial effects even in the absence of light (Fig. 17): (1) disruption of the bacterial cell wall and membrane; (2) binding to FtsZ proteins (homologs of eukaryotic cytoskeletal tubulin) and inhibition of FtsZ protofilament assembly. Curcumin can inhibit the formation of FtsZ and the Z-ring, which are essential for bacterial cytokinesis, thereby preventing bacterial replication; (3) inhibition of the SOS response through interactions DNA and DNA-binding proteins, and (4) suppression of bacterial quorum sensing (QS) through disruption and inhibition of biofilm formation, inhibition of bacterial motility and aggregation behavior, suppression of genes promoting biofilm formation, inhibition of the expression of genes responsible for QS-dependent virulence, and inhibition of bacterial cell growth [61,66]. The phototoxic mechanism of curcumin remains incompletely understood. Its primary antibacterial phototoxic effect is generally attributed to the generation of ROS, which damages lipids, proteins, and DNA, ultimately leading to the eradication of bacterial cells [66]. Huang et al. further reported that curcumin phototoxicity involves direct DNA damage, protein degradation, biofilm eradication, and inhibition of virulence genes (e.g. inlA, hlyA a plcA) in L. monocytogenes [67]. These effects are particularly relevant in hypoxic environments, such as infected tissues, where oxygen availability may limit ROS production [68]. Low or undetectable ROS production in some systems may be explained by Type III photodynamic reactions, which involve mechanisms independent of classical ROS pathways. This may include direct energy transfer to biomolecules or localized physical damage caused by nanomaterials (e.g., sharp edges or mechanical permeabilization) [69]. In addition, the photothermal effect (energy absorption followed by localized heating) may contribute to membrane destabilization and protein denaturation, ultimately causing cell death. Graphene-based nanomaterials possess intrinsic thermal properties, and temperatureinduced membrane damage without ROS generation may account for the observed bactericidal effect [69,70]. Some studies also report oxidative stress induced by glutathione (GSH) oxidation, independent of superoxide generation [71]. cGQDs possess a high surface area and exhibit excellent electronaccepting capabilities due to their unique electronic structure. The properties enable them to capture electrons from excited state molecules such as curcumin, effectively quenching its fluorescence. Upon irradiation, instead of promoting the generation of ROS (e.g., singlet oxygen), electron transfer may result in the formation of reduced species on the cGQDs surface [41,72]. Furthermore, QDs themselves can exert mechanical stress upon binding to bacterial membranes. Due to their nanoscale rigidity and sharp edges, GQDs can disrupt the lipid bilayer or peptidoglycan integrity in bacteria. This can result in membrane perforation or the formation of microcracks, leading to lipid destabilization, leakage of cellular contents, collapse of osmotic balance, and eventually leading to cell death [69]. At higher concentrations, Fig. 17. Possible mechanism of action of cGQDS-PEG-curcumin nanocomposite. K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 16 aggregation of nanomaterials may occur, reducing the effective surface area available for interactions with oxygen and microbial cells. This aggregation can hinder ROS generation despite increased compound concentration. In addition, dense composite suspensions may cause a light-shielding effect, where photon absorption near the surface limits deeper light penetration, thereby reducing ROS production [73]. Another factor that may contribute to undetectable ROS levels is the quenching of ROS prior to detection. Curcumin is known for its strong antioxidant activity. At higher concentrations, it may scavenge free radicals and act as a ROS inhibitor rather than a generator [74]. This antioxidant effect may be further enhanced by PEG, which is known to stabilize curcumin’s antioxidant activity [75]. Additionally, bacteria possess a robust antioxidant defense system. For example, E. coli utilizes catalase, superoxide dismutase, and peroxidase, which can reduce oxidant concentrations and repair damaged lipids [76]. To determine the mechanism of action and localization of the cGQDs-PEG-curcumin nanocomposite, high-resolution fluorescence confocal microscopy was employed. While MIC and MBC values could not be determined after washing the nanocomposite from the bacterial suspension, its fluorescent signal remained visible on the bacterial surface. This observation suggests a localized mechanism of action. Membrane damage was evident, consistent with a mechanism involving localized damage rather than ROS-mediated intracellular effects. resulting in immediate cell destruction. Upon irradiation, cGQDs-PEG-curcumin may also undergo conformational changes that disrupt the membrane at the binding site. It appears that the presence of the nanocomposite in the surface structures of bacteria plays a role in both the protective and phototoxic activity of curcumin [74]. Interestingly, fluorescence images revealed bacterial cell-typedependent differences. Gram-negative bacteria exhibited a more uniform surface fluorescence from cGQDs-PEG-curcumin, whereas Grampositive bacteria showed a stronger and more uneven fluorescence signal, particularly concentrated at the septal region (Figs. 7 and 8). After 24 h of incubation, no intracellular penetration of the nanocomposite was observed. Regions of increased fluorescence likely represent surface aggregates formed through van der Waals forces or electrostatic interactions. If aggregation of the nanocomposite occurs, it may result in an uneven distribution across the bacterial surface rather than a uniform membrane coating. A more homogeneous distribution would be expected if the cGQDs-PEG-curcumin complex interacted exclusively with the phospholipid membrane. In Gram-positive bacteria, the heterogeneous peptidoglycan layer and the presence of teichoic acids may promote localized composite binding. Similarly, outer membrane of Gram-negative bacteria, composed of lipopolysaccharides, may influence the spatial distribution of the nanocomposite. Bacteria undergoing cell division showed enhanced composite binding at the septum, likely due to increased membrane permeability and exposure of nascent cell wall structures. During division, autolysins degrade existing peptidoglycan to allow septum formation, temporarily increasing permeability. The composite’s accumulation in these regions suggests binding to newly synthesized peptidoglycan. Both curcumin and the cGQDs-PEG-curcumin nanocomposite exhibit strong affinity for nucleic acids and amino groups (e.g., lysine residues), supporting their ability to bind bacterial surface proteins [19]. Morphological changes induced by cGQDs-PEG-curcumin and subsequent aPDT were further investigated using SEM (Figs. 9–16). For Gram-positive strains, 5, 10, and 20 µM concentrations were tested, while 20 and 40 µM were used for Gram-negative strains. After two hours of incubation, followed by light activation, significant morphological damage was observed even at the lowest concentrations. Notably, pronounced damage was evident at the highest concentrations even without irradiation, suggesting the inherent antibacterial activity of curcumin. Post-aPDT, SEM images revealed clear signs of cellular disruption, including membrane wrinkling, shrinkage, deformation, surface cracking, cellular debris, and a marked reduction in viable cell numbers. These observations are consistent with previous reports. The study by Mun et al. observed cytoplasmic membrane damage and lysis in a multidrug-resistant S. aureus strain after curcumin exposure [77]. The effects of aPDT using curcumin on Bacillus subtilis were more pronounced with increasing curcumin concentration and irradiation time. Significant damage to the cell wall was observed, including cracks and cell deformation. TEM images revealed blurred cell boundaries, cytoplasm detachment from the cell wall, and changes in nuclear morphology [78]. Morphological changes in Salmonella typhimurium after aPDT with curcumin were demonstrated by Tang et al. SEM images showed that treated bacteria exhibited a wrinkled and irregular surface with cracks, unlike the smooth and intact appearance of the control cells. The study also noted increased membrane permeability and protein damage [79]. The effects of aPDT on S. aureus morphology were very similar. It was found that extended light exposure in the presence of curcumin led to cell wall disruption and bacterial inactivation [80]. 5. Conclusion In this study, we developed a novel nanocomposite consisting of curcumin bound to carboxylated graphene quantum dots functionalized with polyethylene glycol (cGQDs-PEG-curcumin) and demonstrated its potential as a third-generation photosensitizer for antimicrobial photodynamic therapy (aPDT). To the best of our knowledge, this is the first report employing carboxylated graphene quantum dots (cGQDs) as nanocarriers for curcumin in the context of aPDT, offering a new platform for biomedical applications. The integration of curcumin with cGQDs-PEG significantly improved its solubility and bioavailability, enabling effective photodynamic inactivation of bacterial strains, including multidrug-resistant species. The nanocomposite exhibited strong antibacterial activity at low concentrations, with rapid bactericidal effects upon light exposure. Notably, the absence of detectable ROS suggests a non-classical mechanism of action, likely involving direct interaction with and disruption of the bacterial membrane. Highresolution confocal fluorescence microscopy revealed preferential accumulation of the nanocomposite on the bacterial surface, particularly in the septal region, and its efficacy increased with repeated irradiation. Structural damage observed by scanning electron microscopy further supports a mechanism involving membrane rather than intracellular activity. Overall, cGQDs-PEG-curcumin emerges as a highly effective next-generation photosensitizer with the potential to overcome the limitations of conventional antibiotics and ROS-dependent PDT agents. Its ability to function under hypoxic conditions highlights its promise for treating infections in oxygen-deficient environments and addressing the growing challenge of antibiotic resistance. CRediT authorship contribution statement Milan Kol´ aˇ r: Funding acquisition. Kl´ ara ˇ Cepe: Formal analysis. Hanna Dilenko: Writing – original draft, Software, Investigation, Data curation. Hejazi S. M. Hossein: Formal analysis, Writing – review & editing. Hana Kol´ aˇ rov´ a: Visualization, Validation, Project administration, Methodology, Investigation, Funding acquisition. Kateˇ rina Bartoˇ n Tom´ ankov´ a: Writing – original draft, Visualization, Supervision, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Luk´ aˇ s Malina: Formal analysis. Sergii Kalytchuk: Writing – original draft, Formal analysis. Robert Bajgar: Formal analysis, Conceptualization. Ad´ ela Han´ akov´ a: Formal analysis. Renata Veˇ ceˇ rov´ a: Writing – original draft, Validation, Methodology, Formal analysis, Data curation, Conceptualization. Mark´ eta Kolaˇ ríkov´ a: Formal analysis. Lucie V´ alkov´ a: Writing – original draft, Formal analysis, Conceptualization. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 17 the work reported in this paper. Acknowledgements This research was funded by the Ministry of Health of the Czech Republic, grant No. NU21-09-00357 and LM2023050-Czech-BioImaging. The project National Institute of Virology and Bacteriology (Programme EXCELES, ID Project No. LX22NPO5103) - Funded by the European Union - Next Generation EU. S.K. and K.ˇ C. acknowledge support from ERDF/ESF project TECHSCALE (No. CZ.02.01.01/00/ 22_008/0004587). S.K. acknowledges support from the European Union under the REFRESH – Research Excellence For Region Sustainability and High-tech Industries project number CZ.10.03.01/00/22_003/0000048 via the Operational Programme Just Transition. All rights reserved. We sincerely thank Martin Mistrík for his invaluable assistance in confocal microscopy. Data availability Data are available via Zenodo repository at https://doi. org/10.5281/zenodo.17482295. References [1] S. Louca, National antibiotic consumption is strongly related to the prevalence of antibiotic resistance across bacterial clades, iScience 28 (2) (2025), https://doi. org/10.1016/j.isci.2024.111712. [2] S. Bassetti, S. Tschudin-Sutter, A. Egli, M. Osthoff, Optimizing antibiotic therapies to reduce the risk of bacterial resistance, Eur. J. Intern. Med. 99 (2022) 7–12, https://doi.org/10.1016/j.ejim.2022.01.029. [3] K. Patel, R. Panchal, B. Sakariya, et al., Combatting antibiotic resistance by exploring the promise of quorum quenching in targeting bacterial virulence, Microbe (2025). [4] H. Dilenko, K. Bartoˇ n Tom´ ankov´ a, L. V´ alkov´ a, et al., Graphene-Based photodynamic therapy and overcoming cancer resistance mechanisms: a comprehensive review, Int. J. Nanomed. (2024) 5637–5680. [5] Y. Gao, B. Mai, A. Wang, et al., Antimicrobial properties of a new type of photosensitizer derived from phthalocyanine against planktonic and biofilm forms of staphylococcus aureus, Photo Photo Ther. 21 (2018) 316–326, https://doi.org/ 10.1016/j.pdpdt.2018.01.003. [6] Y. Jao, S.J. Ding, C.C. Chen, Antimicrobial photodynamic therapy for the treatment of oral infections: a systematic review, J. Dent. Sci (2023) 1453–1466. [7] K. Varaprasad, N. Sisubalan, T. Jayaramudu, M.M. Yallapu, Nanocurcumin: a new and improved way to fight cancer and infections, in: Nano-Structures and NanoObjects, 40, Elsevier B.V, 2024, https://doi.org/10.1016/j.nanoso.2024.101352. [8] A. Hanakova, K. Bogdanova, K. Tomankova, et al., The application of antimicrobial photodynamic therapy on S. Aureus and E. Coli using porphyrin photosensitizers bound to cyclodextrin, Microbiol Res. 169 (2-3) (2014) 163–170, https://doi.org/ 10.1016/j.micres.2013.07.005. [9] R. Bahrami, M. Pourhajibagher, N. Nikparto, A. Bahador, The impact of antimicrobial photodynamic therapy on pain and oral health-related quality of life: a literature review, J. Dent. Sci. (2024) 1924–1933. [10] S. Marasini, L.G. Leanse, T. Dai, Can microorganisms develop resistance against light based anti-infective agents? Adv. Drug Deliv. Rev. 175 (2021) 113822 https://doi.org/10.1016/J.ADDR.2021.05.032. [11] V. P´ erez-Laguna, I. García-Luque, S. Ballesta, A. Rezusta, Y. Gilaberte, Photodynamic therapy combined with antibiotics or antifungals against microorganisms that cause skin and soft tissue infections: a planktonic and biofilm approach to overcome resistances, Pharmaceuticals (2021). [12] V. P´ erez-Laguna, Y. Gilaberte, M.I. Mill´ an-Lou, et al., A combination of photodynamic therapy and antimicrobial compounds to treat skin and mucosal infections: a systematic review, Photochem. Photobiol. Sci. R. Soc. Chem. 18 (5) (2019) 1020–1029, https://doi.org/10.1039/c8pp00534f. [13] S. Benson, A. Kiang, C. Lochenie, et al., Environmentally sensitive photosensitizers enable targeted photodynamic ablation of Gram-positive antibiotic resistant bacteria, Theranostics 13 (11) (2023) 3814–3825, https://doi.org/10.7150/ thno.84187. [14] L. Tabrizi, R. McGarry, K. Turzanska, et al., Porphyrin-Polymer as a photosensitizer prodrug for antimicrobial photodynamic therapy and biomolecule binding ability, Biomacromolecules (2024), https://doi.org/10.1021/acs.biomac.4c01011. [15] Q. Yao, J. Fan, S. Long, et al., The concept and examples of type-III photosensitizers for cancer photodynamic therapy, Chem 8 (1) (2022) 197–209, https://doi.org/ 10.1016/j.chempr.2021.10.006. [16] M. Nafiujjaman, V. Revuri, H.K. Park, I.K. Kwon, K.J. Cho, Y.K. Lee, Enhanced photodynamic properties of graphene quantum dot conjugated Ce6 nanoparticles for targeted cancer therapy and imaging, Chem. Lett. 45 (8) (2016) 997–999, https://doi.org/10.1246/cl.160388. [17] E. Polat, K. Kang, Natural photosensitizers in antimicrobial photodynamic therapy, Biomedicines 9 (6) (2021), https://doi.org/10.3390/biomedicines9060584. [18] M. Hassnain, L.S. van Hofwegen, H.N. Kaleli, et al., Ultra-Effective Light-Activated antibacterial activity via carboxyl functionalized graphene quantum dots and films, Adv. Funct. Mater. (2025), https://doi.org/10.1002/adfm.202421537. [19] S.C. Gupta, S. Prasad, J.H. Kim, et al., Multitargeting by curcumin as revealed by molecular interaction studies, Nat. Prod. Rep. 28 (12) (2011) 1937–1955, https:// doi.org/10.1039/c1np00051a. [20] B.B. Aggarwal, K.B. Harikumar, Potential therapeutic effects of curcumin, the antiinflammatory agent, against neurodegenerative, cardiovascular, pulmonary, metabolic, autoimmune and neoplastic diseases, Int. J. Biochem. Cell Biol. 41 (1) (2009) 40–59, https://doi.org/10.1016/j.biocel.2008.06.010. [21] R. Kannappan, S.C. Gupta, J.H. Kim, S. Reuter, B.B. Aggarwal, Neuroprotection by spice-derived nutraceuticals: you are what you eat!, Mol. Neurobiol. (2011) 142–159. Humana Press Inc. [22] S. Peng, Z. Li, L. Zou, W. Liu, C. Liu, D.J. McClements, Improving curcumin solubility and bioavailability by encapsulation in saponin-coated curcumin nanoparticles prepared using a simple pH-driven loading method, Food Funct. 9 (3) (2018) 1829–1839, https://doi.org/10.1039/c7fo01814b. [23] R. Bera, B.K. Sahoo, K.S. Ghosh, S. Dasgupta, Studies on the interaction of isoxazolcurcumin with calf thymus DNA, Int J. Biol. Macromol. 42 (1) (2008) 14–21, https://doi.org/10.1016/j.ijbiomac.2007.08.010. [24] T.H. Leu, S.L. Su, Y.C. Chuang, M.C. Maa, Direct inhibitory effect of curcumin on src and focal adhesion kinase activity, Biochem. Pharm. 66 (12) (2003) 2323–2331, https://doi.org/10.1016/j.bcp.2003.08.017. [25] Nafisi S., Adelzadeh M., Norouzi Z., Sarbolouki M.N. Curcumin Binding to DNA and RNA. [26] A. Majhi, G.M. Rahman, S. Panchal, J. Das, Binding of curcumin and its long chain derivatives to the activator binding domain of novel protein kinase c, Bioorg. Med Chem. 18 (4) (2010) 1591–1598, https://doi.org/10.1016/j.bmc.2009.12.075. [27] O. Vajragupta, P. Boonchoong, G.M. Morris, A.J. Olson, Active site binding modes of curcumin in HIV-1 protease and integrase, Bioorg. Med Chem. Lett. 15 (14) (2005) 3364–3368, https://doi.org/10.1016/j.bmcl.2005.05.032. [28] M.H. Karami, M. Abdouss, A. Rahdar, S. Pandey, Graphene quantum dots: background, synthesis methods, and applications as nanocarrier in drug delivery and cancer treatment: an updated review, Inorg. Chem. Commun. 161 (2024) 112032, https://doi.org/10.1016/J.INOCHE.2024.112032. [29] C. Zhao, X. Song, Y. Liu, et al., Synthesis of graphene quantum dots and their applications in drug delivery, J. Nanobiotechnol. BioMed. Cent. 18 (1) (2020), https://doi.org/10.1186/s12951-020-00698-z. [30] S.P. Mohanaraman, R. Chidambaram, A holistic review on red fluorescent graphene quantum dots, its synthesis, unique properties with emphasis on biomedical applications, Heliyon 10 (16) (2024) e35760, https://doi.org/ 10.1016/J.HELIYON.2024.E35760. [31] G. Perini, V. Palmieri, G. Friggeri, A. Augello, M. De Spirito, M. Papi, Carboxylated graphene quantum dots-mediated photothermal therapy enhances drug-membrane permeability, ROS production, and the immune system recruitment on 3D glioblastoma models, Cancer Nanotechnol. 14 (1) (2023), https://doi.org/ 10.1186/s12645-023-00168-9. [32] L. Wang, W. Wang, Y. Wang, et al., The graphene quantum dots gated nanoplatform for Photothermal-Enhanced synergetic tumor therapy, Molecules 29 (3) (2024), https://doi.org/10.3390/molecules29030615. [33] B.Z. Ristic, M.M. Milenkovic, I.R. Dakic, et al., Photodynamic antibacterial effect of graphene quantum dots, Biomaterials 35 (15) (2014) 4428–4435, https://doi.org/ 10.1016/J.BIOMATERIALS.2014.02.014. [34] J. Ge, M. Lan, B. Zhou, et al., A graphene quantum dot photodynamic therapy agent with high singlet oxygen generation, Nat. Commun. 5 (2014), https://doi. org/10.1038/ncomms5596. [35] Y. Yoon, Y.B. Lee, S.K. Kim, et al., Selective liquid-phase extraction of carboxylaterich graphene quantum dots from graphene oxide dispersions, ChemistrySelect 3 (1) (2018) 321–327, https://doi.org/10.1002/slct.201702800. [36] R. Wu, Y. Cao, Z. Chen, J.J. Zhu, Fluorescent graphene quantum dots: properties regulation, sensing applications, and future prospects, Adv. Sens. Energy Mater. 26 (2025) 100140, https://doi.org/10.1016/J.ASEMS.2025.100140. [37] M. Ojeda-Martínez, A.N. P´ erez Martínez, J. El Hamdaoui, et al., Tuning the energy gap of graphene quantum dots functionalized by OH and COOH radicals: first principle study, Mater. Chem. Phys. 311 (2024) 128543, https://doi.org/10.1016/ J.MATCHEMPHYS.2023.128543. [38] D. Du, K. Wang, Y. Wen, Y. Li, Y.Y. Li, Photodynamic graphene quantum dot: reduction condition regulated photoactivity and size dependent efficacy, ACS Appl. Mater. Interfaces 8 (5) (2016) 3287–3294, https://doi.org/10.1021/ acsami.5b11154. [39] R.S. Tade, M.P. More, Emerging application of graphene quantum dots in Photodynamic/Photothermal and hyperthermia therapies for cancer treatment, Nano Biomed. Eng. (2024), https://doi.org/10.26599/nbe.2024.9290083. [40] Isenberg H.D. Clinical Microbiology Procedures Handbook. [41] H. Dilenko, K. Bartoˇ n Tom´ ankov´ a, E. Hinde, L. V´ alkov´ a, M. Kolaˇ ríkov´ a, H. Kol´ aˇ rov´ a, Illuminating the dark: PEGylated carboxylated graphene quantum dots and curcumin in nucleolar activity and PDT-induced DNA damage in cancer, Biomed. Pharmacother. 187 (2025) 118096, https://doi.org/10.1016/j. biopha.2025.118096. [42] F. Feltrin, da, S. Agner, T. Sayer, C. Lona LMF, Curcumin encapsulation in functional PLGA nanoparticles: a promising strategy for cancer therapies, Adv. Colloid Interface Sci. Elsevier B. V. 300 (2022), https://doi.org/10.1016/j. cis.2021.102582. [43] Y.J. Wang, M.H. Pan, A.L. Cheng, et al., Stability of curcumin in buffer solutions and characterization of its degradation products, Pharm. Biomed. Anal. 15 (1997). K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 18 [44] M.L.R. Castillo, Del, E. L´ opez-Tobar, S. Sanchez-Cortes, G. Flores, G.P. Blanch, Stabilization of curcumin against photodegradation by encapsulation in gammacyclodextrin: a study based on chromatographic and spectroscopic (Raman and UV–visible) data, Vib. Spectrosc. 81 (2015) 106–111, https://doi.org/10.1016/j. vibspec.2015.10.008. [45] Z. Zahed, R. Hadi, G. Imanzadeh, et al., Recent advances in fluorescence nanoparticles “quantum dots” as gene delivery system: a review, Int. J. Biol. Macromol. (2024). [46] E. Seyyedi Zadeh, N. Ghanbari, Z. Salehi, et al., Smart pH-responsive magnetic graphene quantum dots nanocarriers for anticancer drug delivery of curcumin, Mater. Chem. Phys. 297 (2023), https://doi.org/10.1016/j. matchemphys.2023.127336. [47] P. Alekseeva, V. Makarov, K. Efendiev, A. Shiryaev, I. Reshetov, V. Loschenov, Devices and methods for dosimetry of personalized photodynamic therapy of tumors: a review on recent trends, Cancers (2024). [48] Y. Cai, T. Chai, W. Nguyen, et al., Phototherapy in cancer treatment: strategies and challenges, Signal Transduct Target Ther 10 (2025). [49] M. G´ orski, J. Nied´ zwiadek, A. Magry´ s, Antibacterial activity of curcumin – a natural phenylpropanoid dimer from the rhizomes of curcuma longa L. And its synergy with antibiotics, Ann. Agric. Environ. Med. 29 (3) (2022) 394–400, https://doi.org/10.26444/aaem/148393. [50] P. Comeau, A. Manso, A systematic evaluation of curcumin concentrations and blue light parameters towards antimicrobial photodynamic therapy against cariogenic microorganisms, Pharmaceutics 15 (12) (2023), https://doi.org/ 10.3390/pharmaceutics15122707. [51] S. Jacob, F.S. Kather, M.A. Morsy, et al., Advances in nanocarrier systems for overcoming formulation challenges of curcumin: current insights, Nanomaterials (2024). [52] A. Karthikeyan, N. Senthil, T. Min, Nanocurcumin: a promising candidate for therapeutic applications, Front Pharmacol (2020). [53] I.F. Almadani, M.F. Almadani, N. AlSawaftah, W.H. Abuwatfa, G.A. Husseini, Nanocarriers responsive to Light—A review, Multidisciplinary Digital Publishing Institute (MDPI), 2024, pp. 827–844. [54] S. Winter, N. Tortik, A. Kubin, B. Krammer, K. Plaetzer, Back to the roots: photodynamic inactivation of bacteria based on water-soluble curcumin bound to polyvinylpyrrolidone as a photosensitizer, Photochem. Photobiol. Sci. 12 (10) (2013) 1795–1802, https://doi.org/10.1039/c3pp50095k. [55] M.K. Morsy, S.Y. Al-Dalain, M.A. Haddad, et al., Curcumin nanoparticles as a natural antioxidant and antimicrobial preservative against foodborne pathogens in processed chicken fingers, Front Sustain Food Syst. 7 (2023), https://doi.org/ 10.3389/fsufs.2023.1267075. [56] C.H.N. Barros, H. Devlin, D.W. Hiebner, S. Vitale, L. Quinn, E. Casey, Enhancing curcumin’s solubility and antibiofilm activity: via silica surface modification, Nanoscale Adv. 2 (4) (2020) 1694–1708, https://doi.org/10.1039/d0na00041h. [57] M. Oves, M.O. Ansari, M.S. Ansari, A. Memi´ c, Graphene@Curcumin-Copper paintable coatings for the prevention of nosocomial microbial infection, Molecules 28 (6) (2023), https://doi.org/10.3390/molecules28062814. [58] F. Bugli, M. Cacaci, V. Palmieri, et al., Curcumin-loaded graphene oxide flakes as an effective antibacterial system against methicillin-resistant staphylococcus aureus, Interface Focus 8 (3) (2018), https://doi.org/10.1098/rsfs.2017.0059. [59] S. Mushtaq, T. Yasin, M. Saleem, T. Dai, M.A. Yameen, Potentiation of antimicrobial photodynamic therapy by Curcumin-loaded graphene quantum dots, Photochem. Photobio. 98 (1) (2022) 202–210, https://doi.org/10.1111/ php.13503. [60] M. Hassnain, L.S. van Hofwegen, H.N. Kaleli, et al., Ultra-Effective Light-Activated antibacterial activity via carboxyl functionalized graphene quantum dots and films, Adv. Funct. Mater. (2025), https://doi.org/10.1002/adfm.202421537. [61] Z. Munir, G. Banche, L. Cavallo, et al., Exploitation of the antibacterial properties of photoactivated curcumin as ‘Green’ tool for food preservation. Int. J. Mol. Sci., 2022. [62] S.Y. Teow, K. Liew, S.A. Ali, A.S.B. Khoo, S.C. Peh, Antibacterial action of curcumin against staphylococcus aureus: a brief review, J. Trop. Med. (2016). [63] M. Wang, Z. Li, Y. Zhang, et al., Interaction with teichoic acids contributes to highly effective antibacterial activity of graphene oxide on Gram-positive bacteria, J. Hazard Mater. 412 (2021), https://doi.org/10.1016/j.jhazmat.2021.125333. [64] S. Yin, J. Liu, Y. Kang, Y. Lin, D. Li, L. Shao, Interactions of nanomaterials with ion channels and related mechanisms, Br. J. Pharmacol. (2019) 3754–3774. [65] Y. Jiao, I.V.J. Wilkinson, E. Christine Pietsch, et al., Iron chelation in the biological activity of curcumin, Free Radic. Biol. Med. 40 (7) (2006) 1152–1160, https://doi. org/10.1016/j.freeradbiomed.2005.11.003. [66] C. Dai, J. Lin, H. Li, et al., The natural product curcumin as an antibacterial agent: current achievements and problems, Antioxidants (2022). [67] J. Huang, B. Chen, H. Li, et al., Enhanced antibacterial and antibiofilm functions of the curcumin-mediated photodynamic inactivation against listeria monocytogenes, Food Control 108 (2020), https://doi.org/10.1016/j.foodcont.2019.106886. [68] A. Etemadi, S.S. Hashemi, N. Chiniforush, Evaluation of the effect of photodynamic therapy with curcumin and riboflavin on implant surface contaminated with aggregatibacter actinomycetemcomitans, Photo Photo Ther. 44 (2023), https:// doi.org/10.1016/j.pdpdt.2023.103833. [69] H. Mohammed, A. Kumar, E. Bekyarova, et al., Antimicrobial mechanisms and effectiveness of graphene and Graphene-Functionalized biomaterials. A scope review, Front. Bio. Eng. Biotechnol. 8 (2020), https://doi.org/10.3389/ fbioe.2020.00465. [70] M.C. Wu, A.R. Deokar, J.H. Liao, P.Y. Shih, Y.C. Ling, Graphene-based photothermal agent for rapid and effective killing of bacteria, ACS Nano 7 (2) (2013) 1281–1290, https://doi.org/10.1021/nn304782d. [71] S. Liu, T.H. Zeng, M. Hofmann, et al., Antibacterial activity of graphite, graphite oxide, graphene oxide, and reduced graphene oxide: membrane and oxidative stress, ACS Nano 5 (9) (2011) 6971–6980, https://doi.org/10.1021/nn202451x. [72] Z. Liang, M. Kang, G.F. Payne, X. Wang, R. Sun, Probing energy and electron transfer mechanisms in fluorescence quenching of biomass carbon quantum dots, ACS Appl. Mater. Interfaces 8 (27) (2016) 17478–17488, https://doi.org/10.1021/ acsami.6b04826. [73] N. Al-Shehaby, H.A. Elshoky, M. Zidan, et al., In vitro localization of modified zinc oxide nanoparticles showing selective anticancer effects against colorectal carcinoma using biophysical techniques, Sci. Rep. 15 (1) (2025), https://doi.org/ 10.1038/s41598-025-00434-3. [74] A. Wolnicka-Glubisz, A. Wisniewska-Becker, Dual action of curcumin as an Antiand Pro-Oxidant from a biophysical perspective, Antioxidants (2023). [75] J.B. Ðokovi´ c, S.M. Savi´ c, J.R. Mitrovi´ c, et al., Curcumin loaded pegylated nanoemulsions designed for maintained antioxidant effects and improved bioavailability: a pilot study on rats, Int J. Mol. Sci. 22 (15) (2021), https://doi. org/10.3390/ijms22157991. [76] M. Fasnacht, N. Polacek, Oxidative stress in bacteria and the central dogma of molecular biology, Front. Mol. Biosci. (2021). [77] S.H. Mun, S.B. Kim, R. Kong, et al., Curcumin reverse methicillin resistance in staphylococcus aureus, Molecules 19 (11) (2014) 18283–18295, https://doi.org/ 10.3390/molecules191118283. [78] L. Dong, J. Qin, L. Tai, et al., Inactivation of bacillus subtilis by Curcumin-Mediated photodynamic technology through inducing oxidative stress response, Microorganisms 10 (4) (2022), https://doi.org/10.3390/ microorganisms10040802. [79] S. Tang, Effect of curcumin mediated photodynamic technology on salmonella typhimurium biofilm and its bactericidal mechanism and application in milk, Mod. Concepts Dev. Agron. 10 (2) (2022), https://doi.org/10.31031/ mcda.2022.10.000731. [80] U. Imaizumi, K. Inaba, A. Kurahashi, et al., Effectiveness of curcumin-based antimicrobial photodynamic therapy against <I>Staphylococcus aureus</I>, J. Oral. Sci. 65 (4) (2023) 23, https://doi.org/10.2334/josnusd.23-0183. K. Bartoˇ n Tom´ ankov´ a et al. Biomedicine & Pharmacotherapy 193 (2025) 118696 19