Impact of PVC microplastics in photodynamic inactivation of Staphylococcus aureus and MRSA

The photodynamic process involves the use of a photosensitizer (PS) molecule, which, at the adequate wavelength of light and presence of oxygen, is able to produce reactive oxygen species (ROS), and singlet oxygen (¹O₂) by photophysical and photochemical mechanisms. The produced reactive species may react with biomolecules and induce to cell death (Robertson et al. 2009). This concept has been extended to various applications, including anticancer therapies (Dolmans et al. 2003) and antimicrobial treatments (Boltes Cecatto et al. 2020; Du et al. 2021), the degradation of chemical contaminants (Gmurek et al. 2017), and inactivation of public health significance, such as diseases vectors (Lima et al. 2022), pathogenic bacteria (Bartolomeu et al. 2017; Ugwuja et al. 2019), and antibiotic-resistant bacteria (ARB) (Sarker & Ahn 2022). In the context of ARB, photodynamic inactivation (PDI) assumes a pivotal role against its spread considering that it is a multitarget inactivation method that relies on oxidative stress (Almeida 2020).

In fact, existing literature presents evidence of the identification of ARB and their genetic components in water (Li et al. 2019), and wastewater treatment plants (Birošová et al. 2014; Rajabi et al. 2022). This underscores the potential significance of suggesting that PDI is a valuable tool for disinfection and tertiary treatment to counteract the prevalence of these microorganisms. Expanding on the environmental applications of PDI, it should be noted that it acts on complex matrices. These matrices often vary in physicochemical parameters, making their quality assessment an essential requirement for selecting the treatment technology (Silva et al. 2022). Different matrices may include specific pollutants that directly influence the mechanisms involved in the treatment train and operational units, presenting a challenge that extends to PDI as well (Almeida et al. 2014; Bartolomeu et al. 2023).

A topic that has been discussed in research on environmental pollution is the presence of microplastics (MPs) in different matrices. Upon disposal, plastic materials undergo various mechanisms leading to the formation of smaller particles measuring less than 5 mm (Zhang et al. 2021). MPs have been detected in various water sources such as seawater, freshwater, urban waterworks, treated tap water, and wastewater (Zhang et al. 2022; Wibuloutai et al. 2023; Yang et al. 2023). A recent study concluded that, along with feed water quality, the efficiency of the treatment plant is one of the major determinants of the vulnerability to contamination of treated water (Oladoja & Unuabonah 2021). The persistence of MPs after undertaking treatment has been demonstrated by numerous studies, which have consistently shown that MPs <0.3 mm often escape in primary and secondary treatment effluents (Blair et al. 2019; Pittura et al. 2021). Regarding tertiary treatment and disinfection, the presence of these contaminants has posed a notable challenge to the efficiency of membrane filtration, largely attributed to fouling mechanisms (Enfrin et al. 2020). Moreover, a recent paper elucidated that MPs caused a negative effect on UV inactivation of ARB (Manoli et al. 2022).

These reports raise questions on how MPs affect the efficiency of PDI against different microorganisms. This matter, to the authors' knowledge, has not been touched upon so far, particularly regarding curcumin as a PS, a natural-based compound that has raised interest in PDI due to its high efficiency and low toxicity, which makes it friendly to environmental applications (Dias et al. 2020). Therefore, the aim of this study was to evaluate the effects of polyvinyl chloride (PVC) MPs on the performance of photodynamic water treatment against Staphylococcus aureus and methicillin-resistant S. aureus, using curcumin as a PS under blue light. Mechanisms were also investigated by photodegradation assays and theoretical molecular modeling.

PVC MPs purchased from Sigma Aldrich MP suspensions were prepared in water, right before PDI assays.

Likewise, curcumin solutions were prepared immediately prior to antimicrobial assays. The powder was purchased from a commercial supplier (PDT Pharma, Brazil) and the master stock consisted of a dilution in ethanol 99.7% (Êxodo^®, Brazil). Work solutions are further described. They were made out of dilutions of the master stock in autoclaved distilled water.

The morphological characteristics of MPs were analyzed by high-resolution scanning electron microscopy (FEG-SEM), with a field emission electron gun (FEG-SEM, Supra 35-VP, Carl Zeiss). The samples were fixed in the substrate on carbon tape, coated with gold/palladium and analyzed at 2.5 kV. Particle size measurement was performed using ImageJ software (version 1.50i). ImageJ software (National Institutes of Health, Bethesda, MD) was used to calculate the PVC MP average diameter.

PDI tests were carried out against two strains of bacteria: S. aureus (ATCC^Ⓡ 25923), and methicillin-resistant S. aureus (clinical isolate), given that previous research reported that they respond differently to PDI in terms of energy requirement (Sammarro Silva et al. 2023; Silva et al. 2023). Stocks were seeded onto Petri dishes containing brain-heart infusion (BHI) agar and incubated at 37 °C for 24 h. Grown colonies were resuspended in BHI broth, then incubated again at 37 °C for 24 h with no agitation. After this incubation period, aliquots were inoculated in BHI and grown to the mid-log phase. Growth was monitored by optical density (600 nm, Bel Photonics spectrophotometer) so that work suspensions of bacteria would be at ∼10⁷–10⁸ colony forming units (CFUs) mL⁻¹ for PDI experiments. Purification was done by centrifuging samples at 3,000 rpm for 15 min (Eppendorf 5,702 centrifuge) and the final pellets were suspended in 1× phosphate saline buffer (PBS, 7.4 pH). These suspensions undertook inactivation tests as detailed further.

After PDI assays, quantification was done by seeding aliquots of serial dilutions (in PBS) onto BHI agar plates, grown in the same conditions as previously stated, so that counts would be available in CFU mL⁻¹.

In a typical experiment, each well was filled up with combinations of purified bacterial suspension (250 μL), MP suspension (250 μL) and PS solution (250 μL) in the desired concentrations. Prior to irradiation, the multi-well plates were incubated for 20 min at room temperature, under static conditions and protected from direct light. The selection concentrations of MP were based on extremely contaminated scenarios (0.25, 0.50, and 1.0 g L⁻¹), following similar research (Manoli et al. 2022) aimed at assessing the effects of the presence of such polymers in photonic disinfection methods.

Irradiation was performed using Biotable^® device, which comprises 24 blue LEDs (emission peak at 450 nm, 41.05 mW cm⁻² at the tested water level). Control experiments that did not contain any PVC (0.00 g L⁻¹) had their respective wells filled up to the same water head, by including the equivalent volume fraction of sterile deionized water. This is also applied for chemical toxicity and photochemical toxicity of MPs investigated in control experiments.

A block of photodegradation assays targeted effects of irradiated PVC microparticles on the photobleaching of the PS molecule. In order to do so, suspensions of PVC in curcumin solution (10 μM; the highest PS concentration used) were irradiated under blue light (LED-based apparatus, emission peak at 450 nm, 21.6 mW cm⁻²). Curcumin was quantified by UV-vis spectroscopy (Varian^® Cary 50 UV-Vis spectrophotometer). The photodegradation percentage was calculated by a decrease in the maximum absorbance peak of the curcumin at 430 nm. Samples were collected under different light doses (0, 5, 10, 25, and 50 J cm⁻²), aiming to provide a photodegradation kinetic profile.

Similarly, photodegradation assays targeting effects on the polymer were carried out. Curcumin (10 μM) and PVC (1.0 g L⁻¹; the highest tested concentration) were irradiated under blue LED light (450 nm, 41.05 mW cm⁻²), at a specific light dose (50 J cm⁻²), based on the UV-vis photodegradation assays. A control group with no irradiation was also considered. The aqueous solutions of the combinations of PVC and the PS were filtered through (7–12 μm) black band quantitative filter paper, aiming to concentrate the particles, and this material was left for drying at room temperature (28 °C, 24 h). The same procedure was performed for the PVC-curcumin suspension without irradiation. Analyses of curcumin powder, PVC, PVC-curcumin, PVC-light, and PVC-curcumin-light combinations were performed by Total Attenuated Fourier Transform Infrared Spectroscopy (ATR-FTIR), using a Cary 630 FTIR spectrophotometer (Agilent^®, USA), spanning the scanning range of 650–4,000/cm with a resolution of 16 cm⁻¹.

Inactivation data were checked for normal distribution using Shapiro–Wilk test (95% confidence interval). Normally distributed data were analyzed by Student's t-test for two samples so that toxicity and phototoxicity could be evaluated. Results on the effect of the presence of MPs in PDI were normalized by log-reduction and grouped into different blocks as per PS concentration and analyzed by one-way analysis of variance (ANOVA) (α = 0.05).

Table 1 provides the p-values resulting from the Student's t-tests of the toxicity and phototoxicity effects of MPs against both tested microorganisms. These are control groups for PVC standalone action. The statistical analyses were based on the residual CFUs corresponding to each specified illumination condition outlined in Table 1. The outcomes suggest no toxicity, except for the highest concentration of MPs against MRSA (contact time of 20 min 18 s). Notably, under light exposure, some degree of microorganism inactivation was detected, though never exceeding 1 log. This result could potentially be attributed to die-off or the photochemical effects to PVC that are further elucidated, given that blue light exposure to MRSA was not significantly toxic, though also led to 0.66 ± 0.08 log-reduction.

For both targets, the presence of MPs in suspension hampered the PDI efficiency progressively. Nonetheless, at the highest concentration of MPs (1 g L⁻¹), results showed a recovery in photodynamic action, though not always reaching (and never surpassing) the full potential demonstrated by the absence of microparticles. The presence of suspended particles in photonic treatments is indeed expected to hinder inactivation efficiency. That is because they act on light scattering, attenuate energy delivery, and enable microorganisms to be shielded from irradiation (Emerick et al. 1999; Christensen & Linden 2003; Manoli et al. 2022). The efficiency of PDI using curcumin against both microorganisms, however, only dropped around 1 log-reduction, except for MRSA in the highest PS concentration. When adding 1 g L⁻¹ PVC MPs to the system, though, we believe additional mechanisms took place to recover inactivation ability, as described in the following topics. This suggests that PDI using curcumin as an efficient approach against bacteria and ARB, even in MP-contaminated scenarios, but investigating details on the mechanisms involved is recommended for future research.

In summary, based on the photodegradation and photoinactivation data, we had expected a competitive process of the ROS produced, as previously reported (Dias et al. 2021). However, depending on PVC concentration, a recovery in PDI efficiency was found, considering results at lower levels of contamination with MPs. This observation may be explained by the formation of Cl. and H. species formed via hydrochlorination process at some specific settings (concentration of the contaminant vs. applied light dose). These reactive species present a high reactive property and may act as an antimicrobial tool in a cooperative way with the conventional ROS formed through a photodynamic process, depending on operational settings and contamination level. In certain conditions, however, they may also increase bleaching of the PS and hinder microbial inactivation, as shown by PDI assays against S. aureus and MRSA.

The (a) spectrum presents a band at 3,508 cm⁻¹ attributed to the presence of the vibrational mode ν(O–H) of the phenolic group. At 1,625 cm⁻¹, the band attributed to ν(C = C) vibrations of the benzene ring bond. Another band at 1,598 cm⁻¹ is attributed to the symmetric stretching vibrations of the benzene ring. At 1,506 cm⁻¹ referring to the vibrational mode of the carbonyl ν(C = O). The band at 1,424 cm⁻¹ olefinic bending vibrations ν(C-H). At 1,268 cm⁻¹ vibrational stretching is attributed to aromaticity ν(C–O), while the band at 1,028 cm⁻¹ refers to stretching vibrations ν(C–O–C). The bands at 954 and 715 cm⁻¹ refer to the trans-CH vibrational modes of the benzoate and cis-CH of the aromatic ring, respectively (Mohan et al. 2012; Garbuio et al. 2022).

Spectrum (b) shows the FTIR for PVC with characteristic bands at 2,967–2,907 cm⁻¹, attributed to vibrational stretching ν(C–H) and at 1,423 cm⁻¹ vibrational stretching of the ν(CH₃) group. The strain at 1,326 cm⁻¹ refers to the vibrational mode ν(CH₂). The band at 1,252 cm⁻¹ can be attributed to vibrational rocking ν(CH). At 961 cm⁻¹, the vibrational mode wagging trans-C–H and at 834 cm⁻¹ vibrational stretching ν(C–Cl), while at 685 cm⁻¹, it is attributed to the vibrational mode cis-CH-wagging (Soman & Kelkar 2009; Kassem et al. 2023). There is no suggestion of curcumin adherence to the polymeric material, as spectrum (c), which refers to the combination PVC-curcumin that does not include the characteristic curcumin bands, though it includes the characteristic bands of the vibrational modes of the PVC polymer. PVC irradiated with a dose of 50 J cm⁻² at 450 nm in the spectrum maintained its characteristic bands, as shown in Figure 6(d). Spectrum (e) showed the appearance of two bands in the regions of 2,366 and 2,340 cm⁻¹, without significant alteration in the other vibrations of PVC. This alteration may be associated with the photodegradation of the PVC polymer via singlet oxygen and free radical oxidation mechanisms (Lee & Li 2021), generated by the PS curcumin when irradiated by a light source at an appropriate wavelength (Dias et al. 2020). These are well-established mechanistic processes in the PDI of microorganisms, and our results suggest that an attack on PVC may also be taking place. At some operational conditions, the effects of this photodegradation may even be favorable for removing contaminants in general and maintaining the performance of PDI.

The formation of ¹O₂ interactions with PVC stereoisomeric molecules was detected by the QTAIM formalism. The bond paths (BPs) showed the formation of interactions between O₂ and the Cl atoms bonded to the C_R atoms (O⋯Cl_R) and to the H atom of the achiral C atom (O⋯H) for the PVC₁ stereoisomer, while in PVC2 observed the formation of an interaction with the Cl atom bonded to the C_S atom (O⋯Cl_S) and with the H atoms of neighboring achiral C atoms (O⋯H). The topological parameters are presented in Table 2. Based on the results of Nakanishi et al. (2008, 2009), all observed interactions are van der Waals in nature, since 0.00 < < 0.01, 0.00 < < 0.04, and 0.000 < < 0.002. Furthermore, the ratio values are less than unity (Bader & Bader 1984; Cremer & Bock 1986), indicating that electrons are depleted in the region between two nuclear attractors, resulting in closed-shell interactions.

Table 2

QTAIM topological parameters for O₂⋯PVC interaction

BCP .	Interaction .	Distance (Å) .	(a.u.) .	(a.u.) .	(a.u.) .	(a.u.) .	(a.u.) .	.
PVC1
P1	OClR	3.469	0.0064	0.0225	0.0047	−0.0038	0.0009	0.8
P2	OH	2.457	0.0109	0.0357	0.0079	−0.0069	0.0010	0.9
P3	OClR	3.474	0.0050	0.0192	0.0040	−0.0033	0.0008	0.8
P4	O⋯ClR	3.429	0.0051	0.0197	0.0041	−0.0033	0.0008	0.8
PVC2
P1	O⋯H	2.516	0.0058	0.0205	0.0043	−0.0035	0.0008	0.8
P2	O⋯ClS	3.391	0.0084	0.0301	0.0068	−0.0060	0.0008	0.9
P3	O⋯H	2.604	0.0080	0.0274	0.0060	−0.0052	0.0008	0.9

BCP .	Interaction .	Distance (Å) .	(a.u.) .	(a.u.) .	(a.u.) .	(a.u.) .	(a.u.) .	.
PVC1
P1	OClR	3.469	0.0064	0.0225	0.0047	−0.0038	0.0009	0.8
P2	OH	2.457	0.0109	0.0357	0.0079	−0.0069	0.0010	0.9
P3	OClR	3.474	0.0050	0.0192	0.0040	−0.0033	0.0008	0.8
P4	O⋯ClR	3.429	0.0051	0.0197	0.0041	−0.0033	0.0008	0.8
PVC2
P1	O⋯H	2.516	0.0058	0.0205	0.0043	−0.0035	0.0008	0.8
P2	O⋯ClS	3.391	0.0084	0.0301	0.0068	−0.0060	0.0008	0.9
P3	O⋯H	2.604	0.0080	0.0274	0.0060	−0.0052	0.0008	0.9

BCP = bond critical point.

View Large

As shown in Table 3, compared to the results obtained for the PVC stereoisomers, ¹O₂ has a very low LUMO energy (E_LUMO) and high electronic affinity ⁠, which may result in electron capture; moreover, ¹O₂ has a low chemical hardness and can become polarized (this justifies the weak interaction with PVC). On the other hand, the polymer has high chemical potential, being able to transfer charge to ¹O₂.

Table 3

Reactivity indices for molecule ¹O₂, and for enantiomeric forms of the PVC obtained at M06-2X/6-311 + +G(d,p) level of theory

Descriptor .	(kcal mol−1) .	PVC1 (kcal mol−1) .	PVC2 (kcal mol−1) .
EHOMO	−209.630	−225.452	−224.430
ELUMO	−68.441	−9.325	−7.613
ΔEH-La	141.190	216.127	216.817
Ionization energy	209.630	225.452	224.430
Electronic affinity	68.441	9.325	7.613
Electronegativity	139.035	117.388	116.021
Chemical potential	−139.035	−117.388	−116.021
Chemical hardness	141.190	216.127	216.817

Descriptor .	(kcal mol−1) .	PVC1 (kcal mol−1) .	PVC2 (kcal mol−1) .
EHOMO	−209.630	−225.452	−224.430
ELUMO	−68.441	−9.325	−7.613
ΔEH-La	141.190	216.127	216.817
Ionization energy	209.630	225.452	224.430
Electronic affinity	68.441	9.325	7.613
Electronegativity	139.035	117.388	116.021
Chemical potential	−139.035	−117.388	−116.021
Chemical hardness	141.190	216.127	216.817

^aΔE_H-L = E_LUMO–E_HOMO.

View Large

The primary aim of this study was to evaluate the impact of PVC MPs on PDI targeting S. aureus and its methicillin-resistant strain (MRSA) in water matrices. Conducted in a controlled in vitro environment, the investigation employed curcumin as the PS and blue light for operation conditions, and high concentrations of MPs to replicate highly contaminated scenarios.

Our findings showed a concentration-dependent hindrance effect exerted by MPs on the efficacy of microorganism inactivation for both S. aureus and its MRSA. This trend was corroborated by photodegradation assays of the PS and was further influenced by the dependence on light energy dosage. Notably, at the highest contamination level, a reversal in photodynamic action was noted, manifesting as an enhancement in inactivation compared to instances with lower PVC concentrations. This recovery was attributed to the formation of supplementary reactive species, a notion validated by the profiles obtained through PS photodegradation assays and supported by theoretical molecular modeling.

Our outcomes highlight PDI as a potential tool for microorganism and ARB control in the contexts of water disinfection and wastewater tertiary treatment, even when confronted with PVC microplastic contamination. Notably, while the presence of MPs does impede efficiency up to a reversible threshold, the processes underpinning photodynamic action extend beyond microorganism inactivation, also applying to the degradation of polymeric contaminants. However, methodological limitations apply to the present paper, inviting future research aimed at elucidating the mechanisms involved in this change from competition to a suggested improvement in PDI efficiency in the presence of PVC as a contaminant. Tests on real environmental matrices and other PS molecules are also encouraged.

The authors are grateful to the 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 (CAPES) (88887.710665/2022-00, 8887.820460/2023-00). The authors also thank Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) (CEPOF 2013/07276-1, 2023/11853–6) and INCT ‘Basic Optics and Applied to Life Sciences’ (FAPESP 2014/50857-8, Conselho Nacional de Desenvolvimento e Pesquisa – CNPq 465360/2014-9). USA agencies also supported this work: GURI -M2303930, CPRIT-RR220054 and CRI-29034.

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

The authors declare there is no conflict.