Realistic environmental exposure to secondary PET microplastics induces biochemical responses in freshwater amphipod Hyalella azteca

ABSTRACT Freshwater environments are especially susceptible to microplastic contamination due to their proximity to urbanised and industrial areas. Also, there is a lack of information about the effects of this pollutant on freshwaters making it difficult the conservation of these environments. Benthic species, such as the freshwater amphipod Hyalella azteca, have been superficially studied so far for evaluation of microplastic pollution. In the present study, we analyzed whether polyethylene terephthalate (PET) microplastics could lead to reduced survival of H. azteca or changes in biochemical markers (SOD, CAT, MDA, and GST) at environmentally relevant concentrations (60 and 600 particles) after 7 d of exposure. The results showed that there was no significant mortality at any of the concentrations tested. The enzyme CAT showed no variation compared to the control group at any of the concentrations. SOD, MDA, and GST were statistically different (p < 0.05). Our study demonstrated that PET MP did not affect the survival of H. azteca at environmentally relevant concentrations. However, changes in biomarkers of oxidative stress may be detected at low level of exposure (60 particles). Although survival is not affected, the macrobenthic invertebrate community may be under threat in environments where there is PET microplastic pollution.


Introduction
Currently, despite the great debates and incentives for sustainable lifestyle behaviours, the production and use of plastic are still unavoidable. Rated by the United Nations Environment Program as the greatest environmental challenge of the twenty-first century, plastic has proved to be considerably a more difficult problem than it was previously expected. One of the main factors for such an issue it is related to plastic were generated by grinding. At the end of grinding, the microplastics were passed through a system of metal sieves to obtain fragments with sizes between 32 and 38 µm. The PET particles were stained with Nile Red fluorescent dye (99% pure, InterLab, Brazil) [21]. A stock solution of Red Nile was prepared at a concentration of 5 g. L −1 diluted in acetone (99%). Immediately before use, the stock solution was diluted to 300 mL in ultrapure water. The solution was added to the PET MP particles, where they remained incubated for 1 h in the dark at room temperature. Then, microplastics were rinsed for 1 min with ultrapure water using a sieve to remove any dye residue, and dried at 50°C.
Initially, a stock suspension of PET MPs was prepared using ultrapure water. This suspension was used to prepare the treatments adopted in the toxicological assays. To confirm the concentration of this stock suspension, we counted, under an optical stereomicroscope, the number of PET MP particles in 10 µL of the suspension. For this, the stock solution was inverted five times and a 10 µL sample was transferred onto a microscope glass slide. Counts were performed under a stereomicroscope (Zeiss Discovery V12) by using a hand counter. This procedure was carried out in triplicate. The average concentration was determined in 59.67 ± 2.08 particles.10µL −1 .
The composition of the fragments was verified by Fourier Transform Infrared Spectroscopy (FTIR). FTIR is a technique applied in the structural investigation of polymeric matrices, in addition to being a well-established, fast, and reliable method of analysis of chemical structures. Thus, it is a very suitable method for analyzing and identifying the composition of microplastics [22]. FTIR spectra were measured in a Bruker spectrometer, Alpha model, in the region of 400-4000 cm −1 , with a standard KBr beamsplitter and high sensitivity DLATGS detector. The spectra were recorded with the ATR (Attenuated Total Reflection) module: ATR Platinum, equipped with a diamond crystal as a reflective element. The morphology of PET microparticles was determined by inverted light microscopy with fluorescence (Leica DMi8).

Exposure of Hyalella azteca to PET microplastics
In the present study, we used adult organisms of H. azteca from a continuous culture in the laboratory. The organisms (n = 5) were placed in glass vials with a metal lid with a capacity of 600 mL and filled with 250 mL of MS culture medium, rich in mineral salts [23], and without aeration. Before the beginning of the experiment, the organisms were submitted to an acclimatisation period, where they were kept without food and monitored for 48 h. The flasks were kept under the photoperiod (16:8 h light:dark) and constant temperature (24 ± 1°C) conditions during the 7-day exposure period. The exposure to PET MPs was via food. An initial solution with a concentration of 5 mg.L −1 of flake fish feed (TetraMin®) and ultrapure water was prepared. In a porcelain container, two concentrations of PET MPs (60 and 600 particles) were mixed with 100 μL of food suspension and oven dried (60°C for 24 h). The containers were placed inside the test bottles to start the exposure period. For the negative control, the porcelain containers contained only food. All treatments and the negative control were performed in triplicate ( Figure 1). Different precautions were taken to avoid contamination by plastic particles present in the environment. Only glass and porcelane glassware was used throught the study. All the test glass flasks and porcelain containers were previously cleaned with 90% acetone. The test solutions were filtered in a glass microfiber filter (47 mm x 0.5 µm) using a vacuum filtration system. Also, to prevent contamination by plastic particles present in the air, the experiments were carried out in a closed room and the flasks were maintained closed with metal lids during the exposure period.

Mortality
The flasks were monitored on days 2, 4, and 7 to record the survival rate of H. azteca organisms. Throughout the experiment, dead organisms were removed from the flasks using a glass pipette. At the end of the exposure period (7 days), the percentage of mortality of organisms compared to the control group was calculated.

Oxidative stress biomarkers
After 7 days of exposure, 3 organisms were used for the preparation of the homogenate (performed in triplicate). Then, the organisms were transferred to 2 mL microtubes containing 600 µL of a 100 mM potassium phosphate buffer pH 7.4 and euthanized in a cold bath for 1 h. The organisms were macerated using a glass rod and, then, centrifuged at 4,000 g for 30 min at 4°C and stored under refrigeration (−80°C). The protein concentration was determined according to the methodology proposed by Bradford [24]. All the biomarkers assessments, including Glutathione S-Transferase (GST), Catalase (CAT), Superoxide dismutase (SOD) and Malondialdehyde (MDA) were monitored through UV-VIS spectroscopy at specific wavelength for each case.

Glutathione S-Transferase (GST)
GST activity was determined as described by Prado et al. [25] with modifications. The assays used 490 µL of 100 mM potassium phosphate buffer (pH 6.5), 490 µL of the mix solution (9.5 mM reduced glutathione (GSH) / 1.0 mM 1-chloro-2,4-Dinitrobenzene (CDNB)), diluted in 100 mM potassium phosphate buffer (pH 6.5) and methanol, respectively, and 20 µL of homogenate. The same solution was prepared for the blank, using a buffer. The assays were carried out in triplicate. GST activity was monitored through the formation of S-(2,4-dinitrophenyl) conjugated glutathione and expressed as μmol CDNB-GSH min −1 . mg protein −1 , by increasing in absorbance at 340 nm for 5 min.

Catalase (CAT)
CAT activity was evaluated using the enzyme assay described by Prado et al. [25]. Assays were performed using 100 mM potassium phosphate buffer (7.0) and 20.0 mM hydrogen peroxide in a 1:1 (v/v) ratio and 20.0 µL of the homogenate. The activity was monitored by consumption of hydrogen peroxide (H 2 O 2 ) resulting in a decline in absorbance at 240 nm for 3 min. The assays were carried out in triplicate. One unit of CAT activity was defined as the consumption of 1 nmol of H 2 O 2 .min −1 .

Superoxide dismutase (SOD)
SOD activity was analyzed by the reaction of pyrogallic acid with the sample, observed at 420 nm, according to Marklund and Marklund [26]. In 2 mL microtubes, 1.3 mL of tris-EDTA buffer (5 mM, pH 8.0), 60 µL of the homogenate, and 75 µL of the pyrogallol solution (15 mM) were added and subsequently homogenised vigorously for 20 s. The assays were incubated for 30 min in the dark at 25°C. After incubation, the oxidation reaction was stopped with the addition of 65 µL of 1M HCl. The same preparation was performed for the blank, using 60 µL of 100 mM potassium phosphate buffer pH 6.5. SOD activity was determined by the ability to inhibit the reduction of pyrogallol by superoxide radicals by 50% expressed as U/SOD. The assays were carried out in triplicate.

Malondialdehyde (MDA)
Lipid peroxidation damage was evaluated through MDA levels, as described by Campos et al. [27], with adaptations. Assays were performed using 0.4% thiobarbituric acid (TBA), diluted in 100 mM potassium phosphate buffer (pH 2.5). In a 10 mL test tube, 2 organisms (H. azteca) were added and macerated with 500 µL of mM potassium phosphate buffer (pH 7.4). Subsequently, 1 mL of 0.4% TBA was added, homogenised, and incubated in a water bath at 95 ± 1°C for 45 min. After being cooled in an ice bath, the samples were centrifuged at 3,000 rpm for 5 min at 25°C and read at a wavelength of 532 nm. The blank solution was prepared from 500 µL of 100 mM potassium phosphate buffer pH 7.4 and 1 mL of 0.4% TBA. The same process was carried out for the standard control, adding 500 µL of 1,1,3,3-tetra ethoxy propane (TEP) 4.5 mM and 1 mL of 0.4% TBA. The assays were carried out in triplicate. The results were expressed as nmol.mL −1 of MDA.

Statistical analysis
Data were expressed as mean ± standard deviation (SD). All statistical analyses were performed using MINITAB software. Data normality was determined by Anderson-Darling test. The significant differences between treatments and control groups were analyzed using Dunnett's test (p < 0.05).

Microplastic characterisation
In the present study, the microplastic particles were produced from a PET bottle by mechanical grinding. The polymeric nature of the particles was confirmed by FTIR analysis, as observed in Figure 2a. In the infrared spectrum, characteristic bands of the polymer are observed. The bands at 1711cm −1 and 1239 cm −1 are attributed to the C = O and C-O stretching of the ester group. The bands at 1016 and 1091 cm −1 are bands indicative of the 1,4-substitution of the benzene ring and the band at 722 cm −1 is the out-ofplane deformation of the carboxylic substituents on the aromatic ring. In the region with the highest wave number, 2700-3200 cm −1 , bands related to the aliphatic CH stretching are observed at ca. 2960 cm −1 and aromatic CH at ca. 3060 cm −1 [28]. The irregular morphology of PET MP was confirmed by inverted fluorescence microscopy images (Figure 2b).

Mortality and oxidative stress on Hyalella azteca
The exposure to PET MPs (32-38 µm) via food for 7 days did not affect significantly the survival of the H. azteca (Table 1). On the other hand, the organisms exposed to PET MP showed alterations in the biomarkers of oxidative stress. CAT did not show alterations in the tested conditions (Figure 3a). SOD was decreased in the highest concentration compared to the control group (Figure 3a). MDA and GST were altered in both treatments (low and high PET). However, while MDA levels increased, GST activity decreased (Figure 3c,d).

Discussion
Considering the growing annual plastic production worldwide [29] and the recent detection of PET particles in surface waters [30][31][32][33][34][35], there is concern about the potential for sediment accumulation and subsequent toxic effects on benthic organisms. In the aquatic environment, the deposition of plastic particles can occur on the surface of the sediment, increasing the exposure of benthic organisms to microplastics, such as H. azteca. This aspect is even more relevant if we consider microplastic as a vector of uncountable pollutants adsorbed on their surface, increasing their toxic potential in the biota [36].
In the present study, we investigated the effects of PET MP (32-38 µm) on mortality and biomarkers of oxidative stress of H. azteca after 7 d exposure via food. The choice of PET particles was due to their environmental relevance. According to Geyer et al. (2017) [29], PET production represents approximately 10% of the total non-fibrous plastics produced annually in the world. It is estimated that an industry of PET bottles can emit 200 g of microplastics per day under normal operating conditions. Moreover, this amount can be 1000 times higher during a heavy rainfall event [37]. Several studies have reported the presence of PET microplastics in water and sediment of freshwater environments [38][39][40][41]. However, data about the effects caused by these particles on freshwater biota, mainly benthic organisms, is still scarce.
In the aquatic environment, fragment-type microplastics can be observed in an infinite variety of shapes, colours, and sizes [39,42,43]. In the present study, we used PET particles of varied morphology with sizes varying between 32 and 38 µm, as observed in Figure 2b.  (32-38 µm) at two concentrations (PET low (60 particles) and PET high (600 particles)) via food for 2, 4, and 7 d. 2d 4d 7d Control 0.0 0.0 0.0 PET low 0.0 6.6 ± 0.0% 6.6 ± 0.0% PET high 0.0 6.6 ± 0.0% 6.6 ± 0.0% The irregular shape is due to the mechanical grinding method used to produce secondary PET particles. Besides, PET microplastics are very representative in freshwaters reaching up to 77.4% of the particles in environmental samples [38]. For this reason, we opted for a realistic environmental condition. For the exposure assays, we adopted two test concentrations (60 e 600 particles).
H. azteca organisms can ingest microplastic present in the environment [17]. Once ingested, these organisms are susceptible to the different effects that these plastic particles can cause. PET MPs seem not to be associated with the mortality of freshwater amphipods. The survival of Gammarus pulex was not affected after exposure to PET MP (10-150 µm, 0.8-4000 particles.mL −1 ) in experiments of chronic exposure (48d) [44]. Similarly, Gammarus fossarum did not show an increase in mortality after 46d exposure to PET films [45]. In the present study, the PET concentrations tested (60 and 600 particles) did not cause significant mortality in H. azteca, determined to be 6.7% for both treatments containing PET (Table 1). On the other hand, enzymatic alterations were observed (Figure 3).
The responses of biochemical biomarkers are endpoints widely used in ecotoxicological studies as indicators of environmental toxicity for different freshwater contaminants [46][47][48]. Studies have demonstrated the ability of microplastics to cause enzymatic changes in freshwater organisms, including fish [49][50][51] and crustaceans [52,53]. However, there is still no information to assist the understanding of the impact of the presence of PET particles in the environment on H. azteca populations.
The increase in antioxidant activity is related to the inability of the redox system in neutralising the reactive oxygen species (ROS) produced by the exposure to microplastics [54]. The results obtained in the present study demonstrated that CAT activity did not show significant changes after exposure to the two concentrations of PET tested (Figure 3a). In contrast, a significant reduction in SOD activity at the highest concentration tested (600 particles). SOD is an important enzyme that, like CAT, is part of the first line of defense against reactive oxygen species (ROS) [55].
The results showed a significant reduction in GST activity (Figure 3d) at both concentrations tested, while MDA levels increased after exposure to PET particles (p < 0.05) (Figure 3c). GST is an important detoxification enzyme capable of removing harmful xenobiotics and endogenous compounds produced from previous detoxification steps [54]. If inhibited, the process of elimination of these compounds by GST can be compromised, increasing the toxic effect of the contaminant on the organism [56]. Inhibition of this enzyme may represent a later stage of response to oxidative stress, leading to the belief that ROS were capable of causing cellular damage [57,58]. Also, the inhibition of antioxidant enzymes may be related to the increase in energetic cost triggered by oxidative stress (GSH) [59]. This hypothesis is supported by the increased activity of MDA, which represents the products of lipid oxidation caused by exposure to the xenobiotic. In addition, MDA also acts as an important indicator of oxidative damage to cell membranes [54] (Figure 3b). Thus, the risk of H. azteca exposure to PET seems to be more associated with the mechanism of lipid peroxidation and the effects of its products. In particular, PET MPs were dose-dependent for SOD and MDA biomarkers. The higher concentration (600 particles) caused more damage than the low concentration (60 particles) (Figure 3 b and c).
Some studies suggest that physical stress may be the main factor related to microplastic toxicity [60,61] when there are no additives in its composition or contaminants that adhere on its surface. Physical stress can occur due to an additional effort to digest inert material and maintain physiological homeostasis [62,63]. Another hypothesis is about microscale abrasions in the internal tissues of these organisms generated by the physical impacts of microplastic particles, which could leave the organism susceptible to other contaminants present in the aquatic environment [62,64].
The results of this study demonstrate that, although environmentally relevant concentrations of PET particles cannot cause the mortality of H. azteca, enzymatic alterations were observed in the exposed individuals. These effects represent a potential risk for the other species of benthic macroinvertebrates and consequently for the aquatic ecosystem since detritivorous organisms such as H. azteca play an important role in the dynamics of these ecosystems.

Conclusion
We investigated the effects of PET MPs (32-38 µm) on biomarkers of oxidative stress of H. azteca in 7-day exposure tests. The present study was the first report addressing oxidative stress by PET particles to H. azteca. We demonstrated that, although significant mortality was not observed, PET plastic particles induced oxidative stress (SOD, MDA, and GST) in amphipods of the species H. azteca after exposure. These findings suggest that the irregularly shaped secondary PET MP may pose a silent threat to benthic invertebrates. Also, the use of biomarkers of oxidative stress can bring relevant sublethal information about microplastic effects on freshwater benthic invertebrates. Our study reinforces that plastic pollution in aquatic environments indeed represents a critical environmental problem for the conservation of freshwater ecosystems.

Data availability statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.