Toxicity of Microplastics and Nanoplastics in Mammalian Systems

Fragmented or otherwise miniaturized plastic materials in the form of micro- or nanoplastics have been of nagging environmental concern. Perturbation of organismal physiology and behavior by micro- and nanoplastics have been widely documented for marine invertebrates. Some of these effects are also manifested by larger marine vertebrates such as fishes. More recently, possible effects of micro- and nanoplastics on mammalian gut microbiota as well as host cellular and metabolic toxicity have been reported in mouse models. Human exposure to micro- and nanoplastics occurs largely through ingestion, as these are found in food or derived from food packaging, but also in a less well-defined manner though inhalation. The pathophysiological consequences of acute and chronic micro- and nanoplastics exposure in the mammalian system, particularly humans, are yet unclear. In this review, we focus on the recent findings related to the potential toxicity and detrimental effects of micro- and nanoplastics as demonstrated in mouse models as well as human cell lines. The prevailing data suggest that micro- and nanoplastics accumulation in mammalian and human tissues would likely have negative, yet unclear long-term consequences. There is a need for cellular and systemic toxicity due to micro- and nanoplastics to be better illuminated, and the underlying mechanisms defined by further work.


Introduction
One of the most prominent classes of non-natural products made by humans that has pervaded Earth's surface environment is plastics, so much so that these durable synthetic organic polymers are heralded as a defining stratigraphic marker for the Anthropocene [1]. Geyer and colleagues recently estimated that 8.3 billion metric tons of virgin plastics have been produced up to the year 2017 [2], and with continuation of current production and waste management practices, about 12 billion tons of plastic waste would be found in landfills and the natural environment by 2050. Plastic wastes are persistent environmental pollutants. Larger pieces of plastic wastes present well-publicized ecological problems in terms of physical entanglement and entrapment [3], physical barriers for food supply [4], and digestive tract congestion. The smaller plastic pieces, particularly their miniaturized forms that are less than 5 mm in size, are generally termed microplastics (MPs) [5]. Plastics that are already small in size to begin with, such as those purposefully manufactured in the form of microbeads in skincare products (primary MPs), or those derived from degradation of larger plastic pieces (secondary MPs), permeate both the terrestrial [6] and the marine [7,8] environments. Plastic particles of less than 1 µm in size are also known as nanoplastics (NPs) [9,10]. These chemically inert MPs/NPs pose significant ecological and health concerns [5] because of their environmental persistence [6,11,12], potential ecotoxicity [13,14], and their ability to act as vectors for chemical pollutants [15][16][17] as well as pathogens [18,19].
The effects of MPs/NPs on mammalian cells and tissues, particularly humans, have remained rather unclear [51,52]. While plastics are generally perceived to pose minimum risk to human, several recent scientific findings, picked up by the popular press, have heightened the worry of possible tissues penetrance and adverse effects of MPs/NPs due to their small sizes. Humans could accumulate MPs/NPs from different food sources [53,54] as well as drinking water [55,56]. Plastic water containers [57,58] and plastic teabags [59] are, perhaps unsurprisingly, common sources for human ingested MPs/NPs. MPs/NPs could also be taken up by inhalation [60]. MPs/NPs have also been detected in human stool samples [61], an indication that the quantity taken in is significantly large. A recent World Health Organization's (WHO) report on "Microplastic in drinking water" indicates that there is not yet proof of harm, but it also calls for more research to be carried out [62].
Could environmental MPs/NPs gain access to cells and tissues and be harmful to humans? Although ecotoxicology data with marine invertebrate indicate that this is so, more barriers and obstacles would likely be encountered by MPs/NPs in order to gain access to cells and tissues of vertebrates compared to simpler invertebrates. Here, we review current results on how MPs/NPs might affect humans by scrutinizing studies done to date on mammalian (mouse) models and human cells. We begin with a quick survey of MP/NP feeding studies done on marine vertebrates, focusing on fishes. A meta-analysis on the effect of MP exposure on fish has been reported by Foley and colleagues in 2018 [63] and the field has also been recently reviewed [64], but several newer reports have now appeared. This quick look would allow some comparison of findings in more ecologically relevant settings with that of laboratory experiments with mice and human cells. Table 1 provides a non-exhaustive summary of recent studies where MP/NP feeding experiments have documented some degree of toxicological or pathological effect observed on fishes. Those that have shown some significant effect are included in this summary, while those that have reported little or no effects are not. The MPs/NPs used in the studies listed in Table 1 are largely polystyrene (PS) or polyethylene (PE) based. An important general phenomenon to note is that toxicological responses typically arise from smaller plastic particles. Larger PS particles at around 100 µm or above were shown not to have any significant effect in a number of studies [65][66][67]. MP/NP feeding can result in behavioral abnormalities in terms of feeding and movement of adults and larvae [35,[68][69][70][71][72][73][74], as well as reproduction in adults [75][76][77]. There is evidence of mother-offspring transfer of NPs [78], and that prenatal exposure of MPs affected early development of the neonates [77].

Toxicity of MPs/NPs in Fishes
In many cases, MPs/NPs were found accumulated in larvae or adult gut [77,[79][80][81][82][83], and in some cases in gill and liver [79]. Histopathology is most prominently observed for these tissues as well [74,75,77,79,[83][84][85]. For the gut, pathological manifestations of MP/NP toxicity include documented changes in gut biomarkers related to epithelial barrier integrity, inflammation, and oxidative stress [83,86], as well as changes in gut microbiota [82,83,86]. In the case of liver, changes in metabolites, key metabolic enzymes, and oxidative stress-induced enzymes occur [49,74,79,81,85,87]. MPs/NPs could be internalized [88], and then cause detectable biomarker changes in blood cells [47,88,89]. In rarer cases, MPs/NPs have also been found in fish brain [68,81], and caused changes in brain appearances [35,68] or showed significantly inhibited acetylcholinesterase (AChE) activity [49,81]. NPs taken up by embryos and larvae have been documented to migrate to various tissues throughout development [70]. Fluorescence red polymer microspheres, (1-5 µm) and mercury individually and in combination • Inhibition of brain acetylcholinesterase (AChE) activity and increase lipid oxidation in brain and muscle • Changes in activity of metabolic enzymes • Interactions and influences on mercury bioaccumulation Barboza et al., 2018 [49] Crucian Carp (Carassius carassius) Amino-modified positively charged PS nanoparticles (52 nm) Trophic transfer to fish from algae through Daphnia. Nanoparticles found in fish brain  Uptake of the nanoparticles by embryos and larvae. Migrated to the gastrointestinal tract, gallbladder, liver, pancreas, heart, and brain throughout development • Decreased heart rate • Altered larval behavior (swimming hypoactivity in exposed larvae) • Maternal-offspring transfer of PS nanoparticles • Delay/defect in swim bladder inflation by exposed F1 larvae  • Fishes with MP have significantly higher lipid peroxidation levels in the brain, gills. and dorsal muscle and increased brain acetylcholinesterase activity Barboza et al., 2019 [95] In line with MPs/NPs activity as carriers or vectors of environmental contaminants, studies using fish cell lines have revealed that, while pristine plastics show no toxicity, those sampled from different islands around the world do [96], and so do those that have been mixed with human pharmaceuticals [97]. MPs/NPs were shown to modulate the toxicities a range of pollutants/toxicants, including phenanthrene [75], mercury [49], cadmium [46][47][48], polychlorinated biphenyls (PCBs) [98], gold ions [99], and the antibiotic roxithromycin [100], in fishes. However, adsorption of toxicants by the plastics could also potentially lower their toxicity, and such is the case for a complex mixture of polycyclic aromatic hydrocarbons (PAHs) [101]. Beyond MP/NP feeding experiments in a laboratory setting, a sampling of wild fishes consumed by humans have indicated that those with MPs found in the gut and other tissues had significantly higher lipid peroxidation levels in the brain, gills, and dorsal muscle and increased brain AChE activity compared to fishes with no MP found in their tissues [95]. These correlations are strongly suggestive of MP/NP uptake being a general stress factor for marine vertebrates.
Overall, despite a large variance in MPs/NPs used, fish models tested, and toxicity parameters examined, there is ample evidence of concentration-dependent acute toxicity as well as chronic effects. Furthermore, environmental toxicants adsorbed onto MPs/NPs would likely change the plastics' toxicity profiles, often with an enhancement of toxicant uptake or an increase in their bioavailability.

Toxicity of MPs/NPs in Mouse Models
In the past three years, a good number of studies have examined the effect of pristine MPs/NPs in mammalian models (largely mouse). These studies are summarized in Table 2 and are broadly recapped below. In mice, ingested MPs/NPs could be found in the gut [102][103][104][105], liver [102,103,105], and kidney [102,105]. Pathological changes to the gut include a reduction in mucus secretion [90], gut barrier dysfunction [104,106], intestinal inflammation [107] and gut microbiota dysbiosis [90,104,106,107]. Liver pathologies documented include inflammation and lipid accumulation or lipid profile changes [90,102,106], as well as changes in the markers of lipid metabolism [90,105,108]. Other metabolic problems noted by omics-type analyses include disorders in energy metabolism [102,105] and bile acid metabolism [104]. On the other hand, a study with mice fed with PS MPs did not reveal any histologically detectable lesions or significant inflammatory responses [109]. A neurobehavioral study on rat fed with NPs also did not detect any significant behavioral changes or abnormality [110].
Recently, Luo and colleagues documented that maternal exposure to PS during gestation causes metabolic disorders in the offspring [106,108]. As in fishes, MPs aggravated the toxicity of an environmental toxicant, organophosphorus flame retardants (OPFRs) [103]. Taken together, the works in mice feeding experiments recapitulated some of MPs/NPs' acute toxicity observed in fish feeding experiments. Such observed toxicities correlated with plastic size [111], concentration [112], and cellular/tissue uptake and accumulation. In general, the degree of MP/NP toxicity observed in mice is less severe than that observed in fishes. One possible reason is that fishes have multiple routes for plastic uptake and accumulation (gut and gills), whereas mice feeding experiments limit uptake through the gastrointestinal route.

Toxicity of MPs/NPs in Human Cells
Could MPs/NPs affect human cells and tissues? There is obviously a lack of toxicity data for humans in vivo at the moment. Several studies have, however, looked at the effect of pristine MPs/NPs on human cells in culture. These works are summarized in Table 3. Not surprisingly, a few of these studies, despite documenting some degree of cellular uptake, found signs of cellular toxicity either absent or insignificant except at very high concentrations of MPs/NPs [109,113,114]. In one case, where polyethylene terephthalate (PET) NPs generated by laser ablation were tested on the human gut adenocarcinoma epithelial line Caco-2, the authors noted a propensity for NP uptakes and crossing of a Caco-2 cells-based intestinal barrier model [113].
A few other studies have documented some degree of cellular toxicity or pathological effect in a range of human cell lines. Prietl and colleagues showed that 20 nm PS NPs are taken up easily by human monocytic cells and are significantly cytotoxic. Larger (100 and 1000 nm) NPs stimulated the secretion of cytokines such as IL-6 and IL-8 from monocytes and macrophages, and could, interestingly, induce a measurable degree respiratory burst in monocytes [115]. Schirinzi and colleagues documented low but measurable degree of reactive oxygen species (ROS) production and induction of cytotoxicity by MPs in T98G and HeLa cells [116]. Wu and colleagues also worked with Caco-2 cells, and reported that, while a low degree of toxicity was observed for PS NPs (at 0.1 and 5 µm), they induced mitochondrial depolarization and inhibited the activity of the toxicant efflux pump, ATP-binding cassette (ABC) transporter, with the latter resulting in increased arsenic toxicity [117]. Hwang et al. worked with a number of cell types of human and mouse origin, and documented cytotoxicity associated with 20 µm PP MPs at high concentrations and ROS induction [118]. The MPs also measurably induced pro-inflammatory cytokines IL-6 and TNF-α from human peripheral blood mononuclear cells (PBMCs), and increased histamine release from mast cell lines [118]. Poma and colleagues found that 100 nm PS NPs stimulated ROS production and induced genotoxic stress and DNA damage as measured with the cytokinesis-block micronucleus (CBMN) assay [119].
Owing to the presence of large amounts of plastic particles in air, terrestrial animals are also exposed to MPs/NPs via inhalation. In this connection, Dong and colleagues found that PS MPs produced some cytotoxic effects, oxidative stress, and inflammatory responses in human lung epithelial cells, and are disruptive of the epithelial cell layer, at least in vitro [120]. Two other groups have recently checked the toxicity of NPs in lung [121] and bronchial [122] cells. Xu and colleagues found that PS NPs (25 and 70 nm) impaired viability, induced cell cycle arrest, and upregulated nuclear factor (NF)-κB as well as some pro-inflammatory cytokines in the human alveolar epithelial line A549 [121]. On the other hand, Lim and colleagues noted that PS NPs are only cytotoxic at high concentrations but induced metabolic changes and endoplasmic reticulum (ER) stress in a human bronchial epithelial cell line [122].
Overall, the experiments with pristine MPs/NPs on human cells reported thus far did not indicate severe cytotoxic or cytostatic effects, but did demonstrate a potential for low to moderate negative effects depending on the cell type, MP/NP sizes, and degree of cellular uptake. Two general and prominently observed phenomena appear to be ROS production and pro-inflammatory responses. In the paragraphs below, we explore how these toxic effects may occur. Table 3. A summary of notable toxicological findings associated with MPs/NPs in human cells. PS, polystyrene; PE, polyethylene; PVC, polyvinylchloride; NPs, nanoplastics (<1 µm); MPs, microplastics.

Human Cell Models
Properties of MPs/NPs Used Cellular Uptake Notes on Toxicological Observations References

Significant Toxicity
Human Peripheral blood monocytic cells

Mechanisms Underlying MPs/NPs' Acute or Chronic Toxicity in Mammalian Cells
In general, extremely high concentration of MPs/NPs are indeed cytotoxic. Cell death could occur via necrotic plasma membrane rupture or some form of programmed cell death. An important point to note on the former, rather non-specific mode of death is the surfactant molecules that are typically associated with most MP/NP preparations. At high concentrations, these would be disruptive to the lipid bilayer of the plasma membrane (PM). Even at moderate levels, these could disrupt important cellular surface structures such as proteoglycans and other extracellular matrix components or hinder cellular signaling processes that require extracellular ligand and cell surface receptor interactions. Therefore, cellular physiology would be affected to varying degrees by plastic associated surfactants, and the documented changes in various transcripts in cells could also be due to this and other processes/factors described below.
The smaller NPs in particular could be taken up with some ease depending on the cell type via endocytosis [123,124]. Endocytosed NPs present a problem for several reasons. Firstly, they could, as per the plasma membrane discussed above, potentially permeabilize the endosomal membranes if present at high concentrations. If this happens, the NPs released into the cytosol could potentially interact with and affect important organelles such as the mitochondria or the nucleus, as well as cellular processes such as mitotic spindle formation and chromosomal migration during cell division. Secondly, MPs/NPs would likely interfere with the trafficking of transport carriers in the cell along the exocytic pathway [125,126], and as such would potentially inhibit the cell surface expression of important signaling receptors or membrane transporters. Thirdly, they are likely to perturb endosomal membrane traffic on which many important cellular processes are dependent, including surface protein turnover and signaling attenuation, as well as retrograde signaling from endosomal compartments. It is unclear if NPs could themselves ever be subjected to inter-compartmental transport efficiently in the endosomal pathway. Even if the NPs could eventually end up in the lysosome, they are unlikely to be readily digested. The accumulation of NPs in late endosome or lysosomes would perturb the degradative functions of these organelles and importantly impair the critical cellular membrane turnover process of macroautophagy [122]. An impairment of autophagic clearance could potentially lead to positive feedback processes that culminate in autophagic cell death. On the other hand, internalized MPs/NPs may also stimulate autophagy. Metallic nanoparticles are known to modulate autophagy [127], and MPs/NPs may speculatively do likewise.
At the very least, these processes would constitute a form of cellular stress. Stresses at the PM and the endo-lysosomes would trigger cellular stress responses. In work done with species of the fresh water flea Daphnia, PS NPs exposure affected growth and reproduction [128], and interestingly resulted in the elevation of AMP activated protein kinase (AMPK), which is an indication of stress response [129]. Perhaps a more general associated phenomenon with regards to cellular stress response appears to be the production of ROS, which was in fact recently identified as the molecular initiating event (MIE) by adverse outcome pathways analysis of reports in the field [130]. ROS production in cells occurs in two general ways: from the mitochondrial electron transport chain (ETC) during routine aerobic respiration or via the oxidative bursts of NADPH oxidases (NOXs) [131]. An increase in ROS from the former could result from mitochondrial function impairment, while the latter is normally a consequence of bacterial invasion, as NOXs are activated by bacterial products and cytokines. All cells are endowed with an evolutionarily conserved innate immunity mechanism, typically functioning against invasion of pathogens or exposure to xenobiotics [132]. However, the components of the innate immune system, such as the Toll-like receptors (TLRs), could also respond to a set of endogenous or secreted molecules collectively known as alarmins, or damage-associated molecular patterns (DAMP) [133,134], and the outcome is what is termed sterile-inflammation, i.e. inflammatory responses without pathogenic infection [135]. In the body, pro-inflammatory cytokines released from such localized inflammations would attract circulating immune cells, and this could worsen the local inflammation, and cause cell and tissue death. NPs has indeed been shown to act as stressors to the innate immune system of fish [88], and this is likely also the case for mammalian (including human) cells. The cellular and tissue invasion and general pathological mechanism of MPs/NPs in mammalian cells is summarized below in Figure 1.

Figure 1.
A schematic diagram illustrating potential (speculative at the moment) cellular mechanisms of MP/NP toxicity. MPs/NPs can be taken up through ingestion and inhalation. MPs/NPs could damage the plasma membrane and impair the gut barrier (left). These could also perturb signaling of cell surface receptors, and alter gene expression in the nucleus. Endocytosed MPs/NPs could also perturb the endocytic pathway function and compromise the endosomal membranes. Stresses arising from the above could activate the cellular innate immune system, with endogenous and secreted damageassociated molecular patterns (DAMP) inducing the innate immunity-mediating toll-like receptors (TLRs). Stresses could induce ROS production from the NADP oxidases (NOXs). Mitochondrial impairment, either by MPs/NPs from endosomes or in response to stresses, could also produce more ROS through impairment in the efficiency of electron transport chain (ETC) processes. MPs/NPs gain access into the circulation if the gut-vascular barrier is compromised or it may speculatively occur by transcytosis, thus reaching other organs. The lung probably has a more direct access to airborne MPs/NPs (right).

Conclusions
MPs/NPs have pervaded the environment and human's exposure and cumulative uptake of these plastics would only increase over time. Currently, it appears that any worries of acute toxicity or severe long-term effect that would lead to significantly enhanced morbidity or mortality are unfounded. However, we still know very little about how MPs/NPs from the environment, be it from a seafood meal or the plastic bottled drink, would affect human health. Clearly, much more research, in terms of both cellular and tissue level pathological mechanisms, as well as on the long-term effects of tissue/organ accumulation, is needed. Plans and collaborative attempts between ecologists and epidemiologists to study bioaccumulation of MPs/NPs in humans via the food chain in various geographical locales would also be necessary.  . These could also perturb signaling of cell surface receptors, and alter gene expression in the nucleus. Endocytosed MPs/NPs could also perturb the endocytic pathway function and compromise the endosomal membranes. Stresses arising from the above could activate the cellular innate immune system, with endogenous and secreted damage-associated molecular patterns (DAMP) inducing the innate immunity-mediating toll-like receptors (TLRs). Stresses could induce ROS production from the NADP oxidases (NOXs). Mitochondrial impairment, either by MPs/NPs from endosomes or in response to stresses, could also produce more ROS through impairment in the efficiency of electron transport chain (ETC) processes. MPs/NPs gain access into the circulation if the gut-vascular barrier is compromised or it may speculatively occur by transcytosis, thus reaching other organs. The lung probably has a more direct access to airborne MPs/NPs (right).

MPs/NPs' Potential Systemic Effect in Humans
MPs/NPs are expected to reach the human gut through consumption of contaminated food materials. Undigested MPs would be largely excreted though fecal matter, but smaller NPs could potential enter the circulation. Ingested MPs/NPs would first encounter the intestinal epithelium. Only unrealistically high concentration of plastics, or those carrying adsorbed toxicants, would likely cause acute impairment of viability and inflammation of the gut lining [51]. However, the effect of persistent presence of inefficiently cleared MPs/NPs in the gut is yet unknown. Gut pathology resulting from MPs/NPs has been widely documented in fishes. The mouse experiments have provided some clear illustration of the consequences of gut toxicity. Should this happen, the gut-vascular barrier could be impaired and MPs/NPs could enter the circulation, where these could gain access to the liver via the portal vein. That this is possible was demonstrated in some of the mouse models [102,103,105]. Long-term accumulation of MPs/NPs in liver tissues and chronic inflammation could lead to liver diseases and metabolic problems. On the other hand, accumulation of MPs/NPs in lung tissues could potentially result in chronic pulmonary disorders. Furthermore, the presence of NPs in brain tissues has been demonstrated in a fish model, as discussed above [68]. It should, however, be noted that it remains to be shown if the MPs/NPs could in fact be found in the brain of experimental mice or human brain samples. For that matter, no cell or tissue accumulation, pathology, or metabolic impairment due to MPs/NPs has been clearly demonstrated for humans to date.
One of the more common findings in the mice studies is gut microbiota dysbiosis [90,104,106,107]. Changes in gut microbiome could result in gustatory dysfunction, thus perturbing physiological homeostasis in general. More importantly, gut microbiota changes have been linked to a range of chronic diseases of other organs, including disease of the kidney [136], cardiovascular system [137], inflammation, and cancer [138], as well as neurological disorders [91,139]. With regards to the latter, gut microbiota dysbiosis could in fact be one of the underlying reasons for behavioral changes in larger animals treated with MPs/NPs. It has also been reported that blood proteins such as albumin and globulin interact with NPs to form protein-plastic complexes [140,141]. Such aggregated protein-plastic complexes, if present in large quantities, could potentially lead to blockage of blood vessels. In addition, while loading of red blood cells (RBCs) with NPs at a low 1:50 ratio did not affect functions of RBCs, a 10-50-fold higher loading showed RBC damages induced by mechanical, osmotic, and oxidative stresses [142]. However, it is difficult to envisage a large acute accumulation of NPs to occur in the human circulation under natural conditions.

Conclusions
MPs/NPs have pervaded the environment and human's exposure and cumulative uptake of these plastics would only increase over time. Currently, it appears that any worries of acute toxicity or severe long-term effect that would lead to significantly enhanced morbidity or mortality are unfounded. However, we still know very little about how MPs/NPs from the environment, be it from a seafood meal or the plastic bottled drink, would affect human health. Clearly, much more research, in terms of both cellular and tissue level pathological mechanisms, as well as on the long-term effects of tissue/organ accumulation, is needed. Plans and collaborative attempts between ecologists and epidemiologists to study bioaccumulation of MPs/NPs in humans via the food chain in various geographical locales would also be necessary.