Propelled by wind, heavy rainfall and tidal currents, MP contamination12,66,67 has spread to remote lakes19,68, rivers14,21,69,70,71, estuarine regions72,73,74, seas41,75, oceans76 and even sea ice77. As a result, a great number of species are currently at risk of exposure and susceptible to ingestion of MPs8,10,25,26,27,28, either through dietary exposure or by transfer along the food chain.
The predominant types of MPs found in aquatic systems are fibres14,15,16,44, microfragments6,7,19,20 and microbeads21. In previous studies, the type of MPs78,79 as well as shape80 and size were suggested to influence the level of toxicity inflicted on fish tissues. To replicate those observations, different types (fibres, fragments and beads), shapes (irregular and regular) and sizes of MPs (1–255 µm) were used in the present study.
Danio rerio were offered free MPs to determine their ability to recognize plastics as inedible particles, as previously suggested by Kim et al.81 Following exposure, all the fish displayed a clear refusal behaviour, suggesting that D. rerio knowingly recognized plastic as inedible particles. Kim et al.81 detailed that capture events in D. rerio had the lowest rates when fish were exposed only to MPs. Most plastic particles were also quickly rejected in a study by Colton et al.82 However, despite recognizing plastic as an inedible element, several authors have documented the presence of MPs in the gastrointestinal tract (GIT) of fishes83,84,85,86,87.
Accidental consumption when foraging on aggregated prey46 has been observed. Additionally, visual cues that resemble prey, such as colour or shape, may enable the ingestion of smaller particles, hindering the distinction between prey and plastic particles22,46,88. Likewise, it has been suggested that odours associated with biofouled plastic debris stimulate foraging behaviour81,89.
In our study, we observed that free smaller fluorescent MPs were occasionally present in the intestinal lumen. It seems that due to their small size (1–5 µm), these fluorescent MPs were unintentionally ingested. In contrast, larger particles of cosmetic MPs were not observed inside the intestine of D. rerio, and accidental ingestion was excluded. Our observations seem to suggest that smaller MPs are more difficult to differentiate from normal prey as previously reported by Critchell & Hoogenboom22 and/or that low-density MPs of a smaller size are more likely to be passively ingested while gulping air46,90.
When MPs were mixed with commercial fish food (diet FA) and fish oil (diet CA), ingestion rates greatly increased. These observations support previous reports that chemical cues resembling prey89 and that higher food concentrations are likely to increase MP ingestion46.
Intestinal retention time, uptake and elimination
The uptake of MP particles was observed in D. rerio fed a single meal of the FA diet. Fluorescent microbeads were detected in the apical and basal surfaces of the enterocytes, inside the goblet cells and in the lamina propria. The internalization of these particles was confirmed by the Z-stack sections. However, despite the uptake of several particles, after 24 h, the MPs had completely cleared from the GITs of D. rerio without translocation to other organs. Penetration of individual particles (up to ~ 70 µm) into the goblet cells has been reported in other species91. Endocytosis of luminal material by goblet cells has been described92, and it has been suggested that some sub-populations of goblet cells may have relatively loose junctions93. In a study by Batel et al., the uptake of a few particles was observed in 7.7% of D. rerio24. However, no further tests were performed to confirm the internalization of the particles. Additionally, the reported incidence was very low.
Conversely, D. rerio fed a single intake of the CA diet presented plastic particles in only the intestinal lumen, and uptake was not observed. In this case, the uptake of microfragments and microfibres seemed unlikely due to their large size. After 24 h, only a microfibre and microfragment remained in the posterior intestine of a single D. rerio, indicating a short retention time for these MPs. Similar observations were described in C. auratus45 and Cyprinodon variegatus80 exposed to 5–200 µm and 6–350 µm particles, respectively.
In our study, the retention rates for MPs and food were similar. However, the retention rate for the fluorescent MPs (1–5 µm) was shorter than that for the cosmetic MPs, as after 24 h, all the fluorescent particles had been completely excreted. Despite their larger size and irregular shape, the cosmetic PE microfragments and microfibres were successfully eliminated in most fish after 24 h. The retention rates for the MPs in our study were similar to those observed by other authors. D. rerio was shown to rapidly excrete MP particles (70 nm–20 µm), reaching a steady state 48 h after exposure48, and C. auratus eliminated 90% of the particles (50–500 µm) to which it was exposed after 33.4 hours45. Similarly, S. aurata showed a retention rate close to zero, as 90% of the fish had cleared the MPs (~ 75 µm) after 24 hours47. Despite the similarities, larger particles are likely to take more time to be eliminated. Seriolella violacea fed MPs (length, 1.2 ± 0.2 mm; diameter, 1.0 ± 0.1 mm) took an average of 10.6 ± 2.5 days to egest the last MPs46. Furthermore, a study by Santos & Jobling performed on Gadus morhua fed a single meal with plastic beads (5,000 µm) and MPs (2000 µm) showed a delay in the evacuation of the 5,000 µm beads when compared to the time it took to egest the 2000 µm MPs94.
Survival rate, weight gain, and feeding behaviour
After sub-chronic dietary exposure to MPs, the survival rates were 100% in all the test groups. For the control group, the survival rate was 95.8%, as a single fish died at the beginning of the sub-chronic experiment. Our results are in line with previous observations in D. rerio exposed to MPs over two95 and 3 weeks48. Conversely, a significant reduction in survival rates was observed in D. rerio fed 10 mg/L PP MPs; the fish in this experiment additionally presented swollen abdomens57.
No significant weight differences were observed between the test groups and the control group. Similar results were reported in Symphysodon aequifasciatus96 and S. aurata exposed to MPs for 30 days97 and 45 days47. Likewise, Acanthochromis polyaccantus exposed to MPs (average 2 mm diameter) for 42 days did not show significant changes in body condition22.
After the second week, the fish fed the CSC diet displayed anxiety-like behaviours, such as anorexia and lethargy, and lightened skin pigmentation. Decreases in feeding and swimming activity were reported in Sebastes schlegelii after exposure to PS MPs98 and Cyprinodon variegatus after exposure to irregular PE MPs80. Lethargy and paling can originate from several factors, such as infections, toxicity, environmental stress, oxygen depletion and starvation99. It was also hypothesized that this decreased response occurred upon repeated exposure to MPs and would be the result of habituation to such particles100. As in another study with D. rerio fed twice a day with nauplii loaded with high concentrations of MPs, there were no observable signs of stress or disease. However, in this case24, the ingested particles were relatively small (1–20 µm), which could explain the absence of signs of distress after the ingestion of multiple meals with MPs.
Particle retention and translocation
For the FSC diet, fluorescent MPs accounted for 0.1% of the total ingested feed, similar to estimations by Jovanović et al.47 For the CSC diet, the average density of MPs used was 2,137 items/m3 (2.49 particles/L). Despite the apparent high number of MP items used, similar or even higher values have been reported in aquatic environments17,69,101.
In addition, it has been reported that most approaches to estimate the numbers of MPs in aquatic ecosystems lead to underestimations23. A common problem with most field studies is that each study uses different mesh sizes, thus having a different cut-off size for the MPs analysed23,102.
After successive meals of the FSC diet, followed by a period of depuration, D. rerio eliminated the particles ingested, as no trace of particles was found in the intestinal tract after the depuration period. In contrast, 50% of the fish fed successive meals of the CSC diet retained an average of 3.3 particles/fish after the end of the experiment. The number of particles found in the intestine after multiple meals is consistent with that found in field studies, in which an average of 0 to 3 particles34,36,37,38,39,40,41,42,43,44 was documented. The low number of particles found in the digestive tract may be indicative of the short residence time of MPs within the GITs of fish. These results also appear to indicate that MPs are unlikely to accumulate within the intestine of fish over successive meals. However, particle size is clearly a determining factor when considering clearance rates for plastic particles. G. morhua fed multiple meals showed a longer retention rate, as the gastric half-life of the beads was substantially increased with particle size94. Our results support the observations by Santos and Jobling94 that MP retention rate seems to increase with increasing particle size and the intake of additional meals.
Nevertheless, it is important to bear in mind that the structure of the digestive tract varies among fish species49. Additionally, other intrinsic (e.g., genetic background, species, age, and physiology) and extrinsic (e.g., habitat, food, type of MPs, and methodology) factors have to be taken into account when discussing clearance rate patterns.
In our study, an average of 1.6 particles/fish was observed in the liver after successive meals of the FSC diet. To validate the process of translocation, the internalization of these particles (up to 1.634 μm) was confirmed by multiple Z-stack sections. The translocation of MP particles to other tissues, such as the liver47,50,51 and muscle37,47, has been previously described. Avio et al. reported the translocation of plastic particles (200–600 μm) from the digestive tract to the liver in M. cephalus50. Likewise, Collard et al. documented the presence of two particles (39–90 μm) in the livers of E. encrasicolus51, and Abbasi et al. detailed the presence of variably sized MP particles (up to 250 µm) in the livers of four fish species captured in the Persian Gulf37. Similarly, Jovanovic et al. observed the presence of < 1 particle (214 ± 288 μm) in the liver of S. aurata47, carefully pointing to the possibility of cross-contamination. The alleged translocation of such large particles is difficult to explain with the current knowledge on translocation pathways for MPs in fish; thus, the plausibility of these reports should be questioned.
Other studies working with smaller particles have also reported the translocation of MPs to the liver48,52. A study exposing O. niloticus to 0.5 µm MPs allegedly observed the translocation of MPs to the liver52; however, the photomicrographs used to support those observations barely show fluorescence in the liver, and the fluorescence in the remaining organs is diffusely spread in the tissues. This is likely to be attributed to leaching of the fluorescent dye and not necessarily the presence of MP particles53. In another study with D. rerio using 5 µm MPs48, the photomicrographic evidence was inaccurate or poorly presented, as previously highlighted in another publication54.
As previously declared by Jovanović et al.47, reports of the translocation of MPs across the fish intestine must be viewed with caution, since the mechanisms of the passage of the plastic material outside of the fish GIT are not yet determined. Two main routes of translocation have been suggested: transcellular and paracellular18,103. The transcellular route involves absorption through the microvillous border to the blood93,103, while the paracellular route occurs through the tight junctions between the cells into the blood104. In mammals, transcellular uptake occurs mostly via M cells in Peyer’s patches and gut-associated lymphoid tissue (GALT)105,106. Fish do not have an organized GALT but instead have lymphoid cells scattered throughout the epithelium and lamina propria and occasional macrophages. Until recently, it was believed that fish lacked M cells107. However, recent studies in salmonids108 and D. rerio109 identified specialized enterocytes with M cell-like activity in the posterior part of the mid-intestine. In fish, these parts of the mid-intestine are the major sites for the uptake of macromolecules and transfer to closely associated intra-epithelial macrophages109. In D. rerio, these cells are identified by the presence of large, supra-nuclear vacuoles109, and it has been suggested that these vacuolated cells deliver luminal contents to scattered immune cells present underneath the epithelial layer110. However, phagocytic activity is not limited to these cells and was also found in regular enterocytes109. The paracellular passage of solid particles through gaps between the enterocytes into the circulatory system has been suggested as the most likely route for MPs, owing to the size range they cover18. In cases of severe inflammation and erosion, the passage of particles through the damaged tissue appears to be facilitated104,111.
MPs could enter the circulatory system by either of these routes and reach the sinusoids through the endothelial fenestrae and the space of Disse. From here, uptake could take place across the basal membrane of the hepatocytes112. The fenestrae vary in size, depending on physiological and pathological conditions, controlling what goes in to or out of the space of Disse and what the hepatocytes are exposed to113. Latex beads of 1 µm and 100 nm were observed within the hepatic sinusoids in both juvenile and adult D. rerio113.
To the best of our knowledge, our study is the first to confirm the internalization of MP particles in the liver, thus validating the translocation of MP particles. In the present study, the translocation of particles was limited to the liver, as no other organs or tissues showed the presence of MPs, despite other authors having described translocation to the muscle37,47.
The digestive tract of fishes is an extension of the external environment, acting as a critical interface between the internal and external environments49 and hence being considered a major route of exposure to MPs. For that reason, intestinal samples were assessed following the guidelines described by Saraiva et al.62.
Our observations revealed the absence of significant lesions in D. rerio from both treatment groups after sub-chronic dietary exposure. Similar results were reported in S. aurata47 and Oncorhynchus mykiss58 fed different types of pristine MPs (0.1 g/kg bodyweight/day) for 45 days and exposed to both pristine and environmentally deployed PS MPs (10 mg/fish/day) for 4 weeks. Similarly, Barbodes gonionotus exposed to PVC fragments (0.2, 0.5 and 1.0 mg/L) for 96 h did not present evident tissue damage114.
Nonetheless, after an exhaustive review of the scientific literature published on the effects of pristine MPs on fish intestines, we realized that the majority of publications present distinct results. Previous intestinal histological changes reported in fish exposed to pristine MPs included cilia defects in adult D. rerio79. As ciliated cells are found in only the intestinal epithelia of lampreys, chondrosteans and dipnoids as well as in early life stages in some teleosts115, they are not present in the intestinal mucosa of most fish species (e.g., D. rerio). In most cases, fish have brush border microvilli instead115, lined by layers of water and mucus49.
Regressive changes, such as erosion and/or ulceration78 of the mucosa, have been reported under the following synonyms: epithelial and villi damage56, cracking of villi, splitting of the enterocytes57 or breakage of the epithelium78. However, such observations must be taken with caution, as erosion and, especially, ulceration are often accompanied by necrosis or haemorrhaging of the mucosa116; these changes are not evident in the photomicrographs offered by any of the aforementioned authors. The detachment of the epithelium from the lamina propria has allegedly been observed55,78,117,118. However, the photomicrographs given to support these observations show a separation between the mucosal epithelium and the lamina propria, which is a common preparation artefact. Other findings, described as a shortening and fusion of the mucosal folds, have also been documented55. Even though persistent toxic damage to the intestinal mucosa and chronic inflammation can produce morphological changes in intestinal folds, such as atrophy and the fusion of adjacent folds, the illusory appearance of these lesions in transverse sections can also be caused by plane-of-section artefacts64. This appears to be the case in the aforementioned study55. Beheading of villi55 has also been reported. Is important to bear in mind that examination of the intestine may be problematic since artefacts due to autolysis occur quickly119. The autolysis of the tips of mucosal folds is a common artefactual finding and often occurs when whole fish are fixed72. Thus, care should be taken when ascribing pathological significance to autolytic changes72.
Vacuolation of enterocytes has been described55,79,117. However, when assessing the vacuolation of the enterocytes in some fish species (i.e., D. rerio), the presence of specialized enterocytes in the posterior segment of the mid-intestine with prominent supranuclear vacuoles has to be taken into account120. A previous study79 has characterized the vacuolation of what looks like a normal enterocyte in the posterior segment of the mid-intestine as a pathological change.
Regarding progressive changes, in our study, the number and size of goblet cells did not show significant differences among the test groups. Likewise, a study conducted on Onchorhyncus mykiss fed PS MPs showed no significant changes in the numbers of goblet cells observed58. Although prior studies95,118 have mentioned mucous hypersecretion, none of the photomicrographs provided showed a significant presence of mucous secretion. In both cases, the authors pointed out to the goblet cells, but mucous hypersecretion was not evident. Thickness of the mucous layer is also known to change by region, being generally greater in the distal sections of the intestine121. Hyperplasia55,117 and hypertrophy of the goblet cells78 have also been reported. Mucous cell hyperplasia is generally associated with sources of persistent irritation, such as parasitism122 and diet119. On the other hand, a decrease in the mucus volume79 and a reduction in the number of goblet cells118 have been reported. Goblet cells are known to vary along the intestinal length121, and for many species of teleosts, the posterior intestine contains the highest concentration of goblet cells123. Fasting has also been shown to reduce the mucosal mass of the intestine124. Another reported finding was hyperplasia of the rodlet cells55. However, no evidence of hyperplasia was given by the photomicrographs available, as rodlet cells can be found in low numbers in healthy tissues. Additionally, care should be taken when ascribing pathological significance to the presence or abundance of these cells because no consistent relation has been established between the numbers of rodlet cells and disease64.
In the present study, inflammation was not observed. Scattered lymphocytes were present in the lamina propria; however, these were considered part of the normal lymphoid tissue. Inflammatory changes56,78 as well as the presence of neutrophils in the intestinal mucosa118 and mast cells at the base of the epithelium79 were observed by other authors. However, these observations are not discernible in the photomicrographs available. The presence of a few leukocytes per se does not necessarily translate to a pathological finding and/or inflammation. In healthy fish, there is a resident population of leukocytes, such as mast cells/eosinophilic granule cells (EGCs), and lymphocytes scattered in the lamina propria62.
Owing to its large blood supply and marked metabolic capacity, the liver is a target organ for toxicants112, while also providing pertinent information about general health and revealing the existence of subclinical background diseases65. In the present study, liver samples were evaluated separately, according to fish gender, following the protocol proposed by Bernet et al.63 Specific sex-related differences are characteristic in adult D. rerio. In reproductively active, adult oviparous females, an upregulated synthesis of the egg yolk protein vitellogenin often causes the hepatocyte cytoplasm to have a mottled, basophilic appearance, with collapsed sinusoids, owing to the hepatocyte enlargement64,124. In contrast, livers from reproductively active males have round eosinophilic hepatocytes with clear vacuoles containing slightly flocculent material and minimal displacement of the nucleus, consistent with glycogen124.
Overall, no circulatory, proliferative, inflammatory or neoplastic changes were noted in our study. After a thorough review of the scientific literature on the effects of pristine MPs on fish livers, we observed that our findings were contrary to those of previous studies, suggesting the occurrence of several changes as a consequence of MP exposure.
Circulatory changes, namely congestion and hyperemia78,117 and haemorrhaging125, have been reported. In a particular case78, the finding described as congestion is likely intravascular eosinophilic proteinaceous fluid. Similar findings were documented by van der Ven et al.126, who identified these changes as an accumulation of vitellogenin in vessels. However, the gender of the animals used by Jabeen et al.78 was not disclosed, and further conclusions cannot be drawn. Overall, when assessing liver congestion and dilated sinusoids, it is important to bear in mind the degree to which the fish was exsanguinated at sacrifice and the amount of care taken to not manually squeeze the liver sample at necropsy64. Liver haemorrhaging125 seems to be characterized by a small number of erythrocytes within the sinusoids, which were likely severed during microtomy. Reports of congestion or dilated sinusoids often are artefacts of tissue collection, preservation or processing65.
Regarding the regressive changes in our study, 22.2% (4/18) of the females presented minimal to mild vacuolation (score: 1–2), while to 27.8% (5/18) presented moderate vacuolation (score: 4). Additionally, 27.8% (5/18) of the males displayed both minimal and mild vacuolation (score: 1–2). A reduction in vacuolation in both sexes appeared to be time-dependent. Glycogen depletion was similarly described in Oryzias latipes exposed to pristine MPs127. The loss of hepatocellular vacuolation is a common response of fish livers to toxicity128. Furthermore, it is a non-specific finding that can occur as a direct effect of intoxication or secondary to decreased body condition caused by inanition, stress or concurrent disease64,124. In our case, vacuolation was not significantly correlated with fish weight, and concurrent diseases were not identified. We believed that the loss of hepatocellular vacuolation might have been caused by stress or even by prolonged exposure to MPs. Paradoxically, toxic exposure can also result in the accumulation of lipids or glycogen in the liver124.
Increased hepatocellular vacuolation117 and vacuolar swelling129 in fish exposed to pristine MPs have also been reported by other authors. However, care must be taken before considering increased hepatocellular vacuolation a pathological change, as it can also be the result of overfeeding an excessively energy-rich diet or lipid peroxidation124. It has been suggested that captive marine teleosts may be particularly predisposed to hepatic lipidosis, as observed in D. labrax used in a study with MPs117, owing to a reduced capacity for hepatocyte peroxisome proliferation coupled with the feeding of artificial diets with high proportions of mono-unsaturated fatty acids124. Lipid peroxidation in fish may also be toxicant-induced65,124. In another study48, lipid droplets in hepatocytes were reported. However, further histochemical techniques were not performed to confirm the lipid origin of the vacuolation. Additionally, due to artefacts in the control photomicrograph, it is difficult to establish a comparison between the control and test groups to identify a possible increase in hepatocellular vacuolation. Apart from lipid and glycogen vacuolation, there are other potential causes of hepatocellular enlargement, such as vacuolar swelling of the endoplasmic reticulum cisternae (hydropic degeneration)124. Vacuolar swelling and hydropic degeneration were allegedly observed in another study78. When comparing the control and test group photomicrographs, it can be observed that cells meant to illustrate hydropic degeneration are also present in the control group photomicrograph. As the magnification differs between the images, there is an illusion of larger vacuoles in the test group photomicrograph. The same applies for the reported vacuolar swelling.
Necrosis was also reportedly observed in the livers of fish exposed to MPs48,117,125. In one study125, necrosis was likely due to handling trauma. Clear spaces without tissue are more likely to be artefacts resulting from focal trauma during tissue collection. In other experiments48,125, there was no evidence of necrosis. Findings classified as hepatocellular necrosis or apoptosis should display cytoplasmic hypereosinophilia with or without condensation; irregular or rounded cytoplasmic margins; nuclear changes, such as pyknosis, karyorrhexis or karyolysis; phagocytosis of necrotic cells or apoptotic bodies; and in the case of necrosis, a potential inflammatory response65. Damaged and aggregated nuclei were documented129. However, these are likely a proliferation of macrophage aggregates. Starvation, ageing, infectious diseases and toxins are all likely to cause proliferation of macrophage aggregates124.
Among the progressive changes, alterations in the hepatocyte morphology and hypertrophy117 were reported. Often, when increased vacuolation results in cytoplasmic enlargement, there is confusion with hepatocellular hypertrophy65. However, hepatocellular hypertrophy should be reserved for describing non-vacuolated cells that are enlarged as a consequence of metabolic enzyme induction, which results in an upregulation of organelles65. In addition, whether due to physiological or toxicological causes, hepatocyte hypertrophy is often accompanied by basophilia124.
Inflammation was also described48,78. In the first study48, signs of inflammation were not evident in any of the provided photomicrographs. In turn, the photomicrographs provided in the second study78 appeared to identify a tubular structure, likely part of the biliary system130 or, less likely, a vascular structure of a normal hepatic stroma. Even though there might be a few leukocytes in the figure provided78, rare leukocytes do not necessarily mean inflammation. When assessing inflammation in the liver, one has to take into account that haemopoietic tissue may be present in the periportal areas of the liver in some fish species131 and that an integral component of inflammation is the infiltration of non-resident leukocytes into the affected site65.
The specific architectural design of fish livers, having only the basal and basolateral aspects of hepatocytes directly exposed to the sinusoidal perfusion, hinders the uptake of chemicals by fish hepatocytes124. This and the lower perfusion rate might help to explain the relative tolerance of fishes to MPs.
Our results show that D. rerio individuals recognize plastic particles as inedible materials but ingest them either when they are mixed with food or fish oil or accidentally when exposed to relatively small plastic particles (1–5 µm). Ingested small plastic microbeads (1–5 µm) and medium PE microfragments (120–255 µm) and fibres (average width and length of 13.67 µm and 1.5 mm, respectively) are unlikely to accumulate in the digestive tract of D. rerio after one or multiple meals, as MPs were almost completely evacuated after 24 h and only a few particles remained in the digestive tract after sub-chronic ingestion. No mortalities or significant effects on body condition were identified after 45 days of feeding with MPs. However, the fish fed medium-sized, irregular PE MPs showed anorexia and lethargy. The ingestion of particles has been reported to cause physical blockage of the intestine, causing a false sense of satiety and interfering with feeding23,34. In the present case, the relatively large particles are thought to have impaired feeding.
To the best of our knowledge, this is the first study to fully demonstrate the uptake and translocation of plastic microbeads to the liver using confocal microscopy. However, the exact route through which MPs reach the liver is still unknown, and future studies are necessary to determine the mechanisms that allow the uptake and translocation of MPs.
Following sub-chronic dietary exposure to pristine MPs, D. rerio did not show any histological lesions in the observed organs. Our results are in contrast to the majority of the scientific literature on the effects of MPs in fish. The differences may be influenced by several elements, such as the species, age, sex, reproductive status of the fish, environment, tested concentrations, size, type, surface chemistry and hydrophobicity of MPs, feeding routine, exposure route, exposure time, number of animals and replicates per treatment group, specimen collection and preparation methods65. However, inaccuracy in the interpretation of the histopathological findings may be the main cause for the disparity observed in the results regarding the effects of pristine MPs on fish. A letter54 written by several veterinary pathologists has highlighted concerns about the recurring problem of inaccurate histopathological data, which is increasingly observed in scientific publications. This situation is especially alarming in cases in which the study conclusions depend heavily on the histopathological results. In addition, such observations will persist in the literature and spawn further misguided research, which is particularly problematic for students and researchers working in fish pathology, expecting to find reliable sources of information in these same publications.
Although pristine MPs per se do not appear to produce imminent damage, most plastics produced are not made entirely of plastic polymers. During the manufacture of plastics, endogenous chemical additives are incorporated into them18. MPs are also very efficient in adsorbing persistent organic pollutants already present in water23. Therefore, further research is needed to properly identify the effects MPs and their associated contaminants may have on animal health and, consequently, public health.