Hydrogen peroxide preoxidation as a strategy for enhanced antimicrobial photodynamic action against methicillin-resistant Staphylococcus aureus

Antibiotic resistance is a topic of major importance in public health (Salam et al. 2023), and it is strongly linked to the environment because of the possible spread of antibiotic-resistant bacteria (ARB) genes. Staphylococcus species are attributed to many human infections and are often prevalent in hospital settings, which also make them an important concern in water and wastewater treatment (Oladipo et al. 2019). Methicillin-resistant, vancomycin intermediate and resistant strains significantly contribute to the disease burden worldwide (Navidinia et al. 2017; Denissen et al. 2022). Their detection in hospital effluents has been reported along with various ARB (Tripathi & Tripathi 2017), but also in municipal wastewater treatment plants, recreational waters, and various environmental matrices (Fogarty et al. 2015; Santos et al. 2020; Rahimi et al. 2021).

Antimicrobial photodynamic treatment (aPDT) is a promising alternative to deal with ARB challenge not only in clinical (Soares et al. 2022; Piksa et al. 2023), but also in environmental applications. One of its major advantages is that aPDT is a multitarget approach that relies on oxidative stress (Almeida 2020). In short, it refers to the use of a photosensitizer (PS) molecule PS when excited by light in the presence of molecular oxygen produces reactive oxygen species (ROS), by the type I mechanism, or singlet oxygen (type II mechanism) (Dąbrowski 2017). aPDT has been widely used in therapies for controlling biofilms and infections in general (Akhtar & Khan 2021; Chen et al. 2022). In the environmental sector, it has shown potentials for vector control, as well as water and wastewater treatment (Lima et al. 2022a, 2022b; Sarker & Ahn 2022).

aPDT has proved to be effective against several targets, as in bacteria, fungi and viruses (Alves-Silva et al. 2023; Gholami et al. 2023; Pourhajibagher & Bahador 2023a, 2023b). However, research is still encouraged for improving efficiency and lowering reliance on energy dose and costly PSs. Preoxidation, often carried out by chlorination, is a strategy in this regard, as it has been used to improve efficiency of operational processes in water and wastewater treatment (Xie et al. 2016). Alternatively to chlorine, H₂O₂ has been applied as a preoxidant, increasing sludge dewaterability (which correlates to cell death), as well as reducing water color and overall microbial contamination (Wei et al. 2019; Silva et al. 2022). A paper in dentistry indicated that preincubation with H₂O₂ and riboflavin was an effective method pre-aPDT against planktonic bacteria responsible for oral infections, and led to microbial reduction in multi-species biofilm (Kunz et al. 2019). Similarly, studies on methylene blue as a PS assisted by H₂O₂ showed improvements in aPDT against bacteria and fungi of clinical concern (Garcez et al. 2011; Yang et al. 2019).

Considering environmental applications of aPDT, we recently reported that no combined effects were obtained against Staphylococcus aureus using curcumin as a PS under an exposure method with no pretreatment with H₂O₂, but rather spiking the oxidant immediately prior to photodynamic inactivation (Sammarro Silva et al. 2023). Results led to antagonistic effects, but the oxidation profile of curcumin in contact with the additive showed stability, which invited further research in H₂O₂ as a preoxidant to aPDT, as explored in peer research with other PS molecules.

Taking this approach with curcumin is a timely effort for aPDT applications in water matrices because it consists of a natural-based PS with low toxicity photodegradation products (Dias et al. 2020; Lima et al. 2022b). As for the additive, peroxidation is a popular decontamination practice using standalone H₂O₂, e.g., in the pharmaceutical industry, clinical environments, aquaculture, agriculture and poultry, food industry, and water treatment (Silva & Sabogal-Paz 2022). Therefore, the aim of this follow-up study was to assess the effects of preoxidation with H₂O₂ in the efficiency of blue light-mediated aPDT using curcumin against methicillin-resistant Staphylococcus aureus (MRSA).

H₂O₂ dilutions were prepared from a 29% stock solution (Synth^®, Brazil). Aliquots from the stock solution were analyzed prior to each experiment. Quantification was performed by the colorimetric method using ferric thiocyanate (Vacu-Vials^® kit, Chemetrics, USA, λ = 470 nm). H₂O₂ dilutions were prepared immediately before each assay. The PS used in this study was synthetic curcumin in powder (PDT Pharma^®, Brazil), and it was applied in solution first diluted in ethanol (Analytical Standards) (Êxodo^® Científica, Brazil), then in distilled water to our work concentrations.

Inactivation assays were performed against MRSA (clinical isolate). A stock suspension previously stored at −20 °C in glycerol (Êxodo^® científica, Brazil) was seeded onto Petri dishes with brain heart infusion (BHI) agar, then incubated at 37 °C for 24 h. Colonies were resuspended in BHI broth, homogenized, then incubated at 37 °C for 18 h. Growth was monitored by optical density (λ = 600 nm, Bel Photonics spectrophotometer) until the mid-log phase, point at which suspensions were purified in sterile phosphate saline buffer (PBS, pH 7.4). Purification was performed by centrifugation at 3,000 rpm for 15 min (Eppendorf 5702 centrifuge). After inactivation assays, viable colonies were quantified by serial dilutions in PBS, spread onto BHI agar and grown for 18 h at 37 °C. Final counts were considered in CFU mL⁻¹ and inactivation values were calculated as described further.

The experimental setup comprised two stages: incubation and irradiation, which is the standard protocol for aPDT. Enhanced treatments included the presence of H₂O₂ during incubation with the PS, which may be considered a preoxidation step. All aPDT assays were carried out in 24-well plates illuminated by a light-emitting diode (LED) device equipped with 24 blue LEDs (Biotable^®, λ = 450 nm), leading to an average irradiance of 41.05 mW cm⁻² at the tested water level. Applied fluence was 50 J cm⁻², adapted from peer research against MRSA (Almeida et al. 2017). The emission spectrum of the Biotable^® device is provided in the supplementary file. In order to maintain the water head, sample volumes of equal proportions would always lead to a total of 750 μL, accounting for the combinations of H₂O₂, curcumin, MRSA suspensions, and autoclaved distilled water when necessary. Selected concentrations for mapping H₂O₂ effect were defined according to previous studies in combined physicochemical treatment (Sammarro Silva et al. 2022), and combined aPDT with curcumin, specifically (Sammarro Silva et al. 2023).

Oxidation tests were also carried out in order to identify the contribution of H₂O₂ action as an individual disinfection method (as per the preoxidation represented in Figure 1). Experimental setup was the same as for sequential treatments and undertook 30 min exposure time of bacteria to a range of H₂O₂ concentrations. Identical oxidation assays were performed with an additional contact time, in order to compensate for possible prolonged exposure in enhanced aPDT tests (preoxidation + illumination time), as no quenching of residual H₂O₂ was done. Selected concentrations for standalone oxidation test were result-oriented, based on the obtained performances of H₂O₂-assisted aPDT. These blocks of experiments were performed in absence of direct light, also in multi-well plates that were kept incubated in the dark at room temperature and static conditions during contact time.

Aliquots of MRSA, grown and purified identically to the inactivation assays, were incubated for 30 min in different test groups: internalization (IT), which comprised bacteria suspended in curcumin (10 μM) and water, and curcumin and H₂O₂ (0.03%), and PS control, that referred to a curcumin solution (10 μM) in distilled water. Combinations considered equal parts of each medium (1 mL) and proportionate dilutions, as the PS solution was originally prepared at 3 × .

After the IT period, the samples were centrifuged (Eppendorf centrifuge 5702) at 3,000 rpm for 15 min. Two mL of the supernatant was added to a quartz cuvette (10 mm optical path) and subjected to UV–Vis absorption spectrophotometry (Varian Cary Bio 50).

Possible peroxidation of the PS was assessed by monitoring the absorbance at 430 nm of different combinations of curcumin and H₂O₂ at various concentrations. This wavelength was chosen based on ultraviolet–visible (UV–Vis) absorbance spectroscopy of curcumin solutions, described in the supplementary file. The exposure time was selected in order to simulate preincubation performed in enhanced aPDT assays (30 min), as well as prolonged contact time, accounting for irradiation period. Test concentrations of the additive were selected in an exploratory manner, starting out from the lowest (0.01%), the highest (0.30%), and a few intermediate ones. Absorption spectrophotometry was performed in a 10 mm optical path (Varian Cary Bio 50 spectrophotometer). The experiment was performed for n = 3, and results were analyzed by t-test (α = 0.05) against a given mean value (standalone PS).

As for reports on conventional aPDT against MRSA, studies have shown that it is more resistant to inactivation than the methicillin-sensitive strain. By using an edible die as PS (sodium copper chlorophyllin) under red light (30 J cm⁻²), a study found that efficiency was concentration-dependent regarding the PS, but even higher concentrations (20 μM, for instance) did not reach complete MRSA inactivation (Caires et al. 2020). Similarly, a study on phototoxic effects of curcumin against bacteria screened concentrations and light doses over the sensitive strain, in order to define experimental conditions for the photokilling of MRSA, and only the highest tested concentration of the PS (20 μM) was effective against the resistant strain, at a 37.5 J cm⁻² fluence of blue light (Ribeiro et al. 2013).

Regarding improved methods of aPDT, an in vitro study including potassium iodide and iodopovidone as additives to a porphyrin formulation led to substantial increase in MRSA inactivation (Braz et al. 2020). In that paper, the additives were incubated with the PS for 10 min prior to photodynamic assays, inviting for research in sequential aPDT. A similar study in photodynamic inactivation of MRSA achieved an increase in efficiency in 5-log₁₀, under 100 J cm⁻² (laser, near infrared light), applying indocyanine green (ICG) as a PS. It should be noted that, in this paper on ICG, the tested H₂O₂ concentration was 0.1% (Wong et al. 2018), higher than the breakpoint we found for curcumin. This not only substantiates the positive impact of pretreatment to aPDT, but also endorses the importance of screening tests for optimizing operational conditions. These will lead to a reduced dependence on light and the PS concentration, based on the selection of parameters considering the three independent factors aimed at minimum microorganism survival. Overall, the improved inactivation followed by deleterious effect found in our present study point H₂O₂ preoxidation and aPDT as a promising method for disinfection and tertiary treatment against ARB.

Details of H₂O₂ contribution for 30 min incubation and extended contact time are laid out in Table 1 along with their respective SFs. As expected, prolonged exposure contributed to an increase in bacterial inactivation, but, as we worked with low H₂O₂ concentrations, this reduction is not very pronounced.

When considering oxidation performance as a single factor, MRSA log-reductions met expectations for standalone liquid H₂O₂, because this oxidant requires high CT values for drops in microbial load (Silva & Sabogal-Paz 2021). Against MRSA, most published research relies on different application methods such as dry mist (AHP), vaporized H₂O₂ (VHP) and peroxygen-based formulations applied as either AHP or VHP for surface disinfection. A study testing inactivation efficiency using contaminated carrier disks, for instance, targeted MRSA using 5% H₂O₂ AHP for 2.5 h (Piskin et al. 2011). Also applying AHP, similar research tested swabs from high-touch surfaces in clinical environments with patients with known multi-resistant organisms colonization following manufacturer's instructions for Deprox AHP, but operational details are not available, though there was a decrease in rates of contaminated surfaces after disinfection (McKew et al. 2021). Similarly, an automated VHP system was tested at 5, 10, and 35% (w w⁻¹) against MRSA for 40 min exposure followed by 200 min dwell time, also for surface disinfection (Murdoch et al. 2016). It is therefore challenging to extrapolate such results to our experimental content.

Though individual H₂O₂ action is low, SFs listed in Table 1 point it to a promising preoxidant to aPDT and quantify data on 0.04% as the breakpoint concentration for favorable inactivation, as shown in Figure 2. Such favorable results for sequential action stand out even more when we consider the very low concentrations selected for H₂O₂. In environmental applications, these may be considered insignificant, based, for instance, on the exemption in the list for tolerance in terms monitoring and quenching residuals in food decontamination (as long as application is ≤1%) (USEPA 2002).

Similarly, curcumin is considered an environmental-friendly PS that will be consumed in the course of aPDT action, and, when applied in the very low concentration proposed for the H₂O₂-enhanced treatment, does not lead to any increase in color. Moreover, its possible residual byproducts are safe, as shown in aquatic ecotoxicity assessments previously published, where authors showed that any photobleaching intermediates do not present toxicity potential against green algae, fish, nor Daphnia (Lima et al. 2022b), also confirmed by bioassays performed against non-target organisms (Venturini et al. 2020), indicating that curcumin-driven aPDT does not raise ecotoxicity issues.

Assuming that there was no increase in IT in H₂O₂-pretreated aPDT, we may infer that synergistic effects were obtained based on preoxidation as a single factor, as previously shown, followed by an improved photodynamic inactivation on remaining organisms, possibly more fragile. We believe that this synergistic effect obtained in sequential treatment is promising for environmental applications, where challenging targets persist through the treatment train, but also interesting future research in aPDT in general, e.g., control of biofilms and infections.

Quenching of H₂O₂ radicals is a topic that must be considered to determine whether there are any residuals to be quenched by an external agent or the substrate itself. As shown by similar studies in sequential aPDT, their presence may reduce the production of other radicals (Garcez et al. 2011; Garcez & Hamblin 2017). Also, organic matter in solution acts as oxidant demand at the first minutes of exposure (Freitas et al. 2021; Silva et al. 2022).

Table 2 displays data on curcumin peroxidation monitoring at 430 nm, considering different exposure times and concentrations of H₂O₂. In general, no significant differences in means were found against the PS under peroxidation stress, which suggests that the presence of H₂O₂ in solution did not damage curcumin. The only situation in which p < 0.05 was at 0.04% H₂O₂ during prolonged oxidation, which we may attribute to various reasons such as operational error, curcumin solubility, or randomness. These may be considered positive results, because they suggest that the presence of the oxidant during incubation does not impair the integrity of the PS molecule, and, therefore, the performance of aPDT, and, when looking at the oxidant demand, it does not compete with bacteria.

Results found for preoxidation with H₂O₂ led to promising takes for disinfection applications and infection control, as it even caused a deleterious effect, as shown in Figure 2. This absence of microorganisms, however, must consider inherent limitations in the detection method of viable bacteria. The quantification protocol requires extrapolations for determining the microbial concentration in CFU mL⁻¹. Future work to delineate optimal operational conditions may shed light onto details for the H₂O₂ concentration dependence and include additional quantifiers for bacterial viability.

Further research must comprise information on survival of pathogens from different core groups (e.g., viruses, protozoa, helminths, etc.), so that infection risk will be accurately calculated (Mraz et al. 2021). Along with the target challenge, water quality plays a key role in oxidation performance (Silva et al. 2022), and will affect the efficiency of both the pretreatment and aPDT. Likewise, removing biofilms and preventing biofilm formation also represent challenges to disinfection. Given recent studies in aPDT for clinical applications against biofilms (Akhtar & Khan 2021; Alagha & Gülsoy 2023), there are promising perspectives to it in environmental applications that may be further assessed in enhanced treatment (Yang et al. 2019). This also brings perspectives toward kinetics studies, modeling and in silico assessments.

As for aPDT as a single factor, though curcumin is known as an environmentally safe PS molecule (Dias et al. 2020; Lima et al. 2022b), it is crucial that no residuals or photoproducts are present in treated samples, particularly considering the organic input. Future work on optimal operational conditions must therefore consider mineralization of the PS. Including the pretreatment step with H₂O₂ may lower required concentrations of the PS (or energy input) and this should be considered a potential avenue for work aimed at more sustainable technologies in advanced water and wastewater treatment. These may even include tests on full spectrum of visible light- or solar-based systems.

The stated objective of this study was to provide a follow-up on the effects of H₂O₂ as a pretreatment to aPDT against bacteria. Though recent research showed antagonistic effects in combined aPDT, our newly proposed method led to cooperative effects. The method consisted of preoxidation by incubating the PS along with the target (planktonic MRSA) prior to photodynamic assays, carried out using curcumin under blue light. Here, we showed that it acts in a synergistic manner, though leading to some slight impairing effects in lower doses, but with a breakpoint substantially favorable to aPDT at 0.04% H₂O₂, followed by complete bacterial inactivation, a deleterious effect that is really promising for advanced treatment in water and wastewater matrices.

The proposed approach for sequential treatment (tested in vitro at this point) relies on environmental-friendly solutions for both applied chemicals, and their photoproducts. Also, describing an increased efficiency of aPDT invites research in optimal conditions for the assisted technology, aiming to reduce reliance on light dose, and its associated costs, as well as the PS. Our results, therefore, describe potentials for using H₂O₂ preoxidation as a strategy to improve aPDT performance and tackle the issue of ARB in environmental matrices.

This work was supported directly and indirectly by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) (CEPOF 2013/07276-1) and INCT ‘Basic Optics and Applied to Life Sciences’ (FAPESP 2014/50857-8, CNPq 465360/2014-9), and by Empresa Brasileira de Pesquisa e Inovação Industrial (EMBRAPII). The authors are also grateful to Fundação de Amparo à Pesquisa do Estado de Goiás (FAPEG) (201810267001556 and Inovação, Desenvolvimento e Sustentabilidade: Estreitamento entre Universidade e Setor Produtivo no Estado de Goiás Convênio para pesquisa, desenvolvimento e inovação – PD&I 07/2020), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (88887.710665/2022-00, 8887.820460/2023-00), Cancer Prevention and Research Institute of Texas (CPRIT, M20301556), Governor's University Research Initiative (GURI, M230930), and CRI (02-292034).

MRSA isolates were obtained from samples of a previous clinical study on pharyngotonsillitis carried out with permission from the ethics committee CAAE number: 83082018.4.0000.8148 (Brazil).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.