Micro- and nanoplastics (MNPs) and their potential toxicological outcomes: state of science, knowledge gaps and research needs

Abstract

Plastic waste has been produced at a rapidly growing rate over the past several decades. The environmental impacts of plastic waste on marine and terrestrial ecosystems have been recognized for years. Recently, researchers found that micro- and nanoplastics (MNPs), micron (100 nm – 5 mm) and nanometer (1 – 100 nm) scale particles and fibers produced by degradation and fragmentation of plastic waste in the environment, have become an important emerging environmental and food chain contaminant with uncertain consequences for human health. This review provides a comprehensive summary of recent findings from studies of potential toxicity and adverse health impacts of MNPs in terrestrial mammals, including studies in both in vitro cellular and in vivo mammalian models. Also reviewed here are recently released biomonitoring studies that have characterized the bioaccumulation, biodistribution, and excretion of MNPs in humans. The majority MNPs in the environment to which humans are most likely to be exposed, are of irregular shapes, varied sizes, and mixed compositions, and are defined as secondary MNPs. However, the MNPs used in most toxicity studies to date were commercially available primary MNPs of polystyrene (PS), polyethylene (PE), polyvinyl chloride (PVC), polyethylene terephthalate (PET), and other polymers. The emerging in vitro and in vivo evidence reviewed here suggests that MNP toxicity and bioactivity are largely determined by MNP particle physico-chemical characteristics, including size, shape, polymer type, and surface properties. For human exposure, MNPs have been identified in human blood, urine, feces, and placenta, which pose potential health risks. The evidence to date suggests that the mechanisms underlying MNP toxicity at the cellular level are primarily driven by oxidative stress. Nonetheless, large knowledge gaps in our understanding of MNP toxicity and the potential health impacts of MNP exposures still exist and much further study is needed to bridge those gaps. This includes human population exposure studies to determine the environmentally relevant MNP polymers and exposure concentrations and durations for toxicity studies, as well as toxicity studies employing environmentally relevant MNPs, with surface chemistries and other physico-chemical properties consistent with MNP particles in the environment. It is especially important to obtain comprehensive toxicological data for these MNPs to understand the range and extent of potential adverse impacts of microplastic pollutants on humans and other organisms.

1. Introduction

Roughly 8.3 billion metric tons of plastics were produced by 2015, resulting in the production of 6300 metric tons of plastic waste [1]. The world is generating plastic waste at twice the rate that it was two decades ago, with most entering the environment. With the continued exponential increase in the production of plastics without effective plastic waste management strategies, an estimated 12 billion metric tons of plastic waste will be deposited in landfills and natural environments by 2050 [1, 2]. The COVID-19 pandemic has dramatically increased single-use plastic waste production due to the enormous surge in consumption of personal protective equipment (PPE) and packaging plastics [3]. The continuing increase in plastic waste generation, coupled with increasing evidence of environmental contamination by small plastic particles and fibers generated from that plastic waste, has resulted in growing concern in both the environmental protection and public health communities.

Micro- and nanoplastics (MNPs) are defined as plastic particles and fibers with sizes ranging from nanometers (≥ 1 nm) to micrometers (≤ 5 mm) [4]. Specifically, microplastics are characterized by particle sizes between 100 nm and 5 mm, while nanoplastics fall within the size range of 1 to 100 nm. MNPs can be categorized as either primary or secondary. Primary MNPs are particles that are intentionally manufactured for commercial applications such as microbeads in personal care products. Secondary MNPs refer to fragments produced from the degradation of large plastic debris or primary MNPs via exposure to various degrading factors, including UV radiation from sunlight, physical abrasion, thermal stress, oxygen stress and/or biological processes [5-11]. These include particles generated by photo-oxidative degradation and mechanical forces acting on plastic waste deposited in terrestrial and aquatic environments, municipal incineration of plastic trash, as well as those generated during the use of plastic products, such as microfibers generated from laundering of synthetic fabrics, tire dust created through friction and abrasion of tires moving on the road, and particles released from artificial turf fields and plastics-based paints, coatings and toner based printing equipment and 3D printers [2, 12-19]. These abiotic and biotic processes are collectively referred to as weathering of MNPs [20]. Different types of MNPs present various morphologies and compositions, with high particle heterogenicity compared to engineered nanomaterials [7]. The physico-chemical characteristics of MNPs can impact their transport, distribution, and fate in the environment [21]. In addition, a variety of chemical additives are often added to plastics during manufacturing to improve the properties and performance of the final material. These additives can potentially be released from MNPs to pose an additional chemical hazard [22]. Furthermore, because of their capacity to adsorb environmental pollutants (EPs), including persistent organic pollutants, trace metals, microorganisms and leachates of additives, MNPs can serve as vectors of toxic substances, possibly potentiating their toxicity [22-25].

An enormous amount of persistent and slowly degrading plastic debris has been released into marine environments over the past several decades and is becoming increasingly problematic [5]. Ultimately all plastics will be fragmented into MNPs, causing increasing threats to marine biota and ecosystems. In recent years, various studies have reported the adverse effects of MNPs on marine ecosystems; these studies have been comprehensively reviewed [26]. Uptake and accumulation of MNPs in marine organisms induced a variety of toxicological effects, including oxidative stress, energy and metabolic disruption, immune dysfunction, neurotoxicity, and physiological and behavioral alterations [27-32]. MNPs can enter the human body through ingestion of MNP-contaminated water or food products, inhalation of airborne particles and fibers, and dermal contact with MNP-containing products, potentially posing a direct threat to human health [33]. Recently, an increasing number of studies have employed mammalian animal and cellular models to assess the health impacts of MNP exposure on terrestrial mammals and humans. For example, MNP exposure studies in rodents have shown that ingested or inhaled MNPs can cross physiological and biological barriers to reach the circulatory system and secondary organs, suggesting the potential for absorption, distribution, accumulation and toxicity of MNPs [34-39]. At the cellular level, internalized MNP particles can cause lysosomal damage, oxidative stress, inflammation, mitochondrial dysfunction, and cytoskeletal perturbation, leading to cell injury and death [40-44].

With the rapid increase of scientific publications evaluating the impacts of MNPs on mammalian systems, an integrated and comprehensive review is urgently needed to evaluate the current understanding of the potentially detrimental effects of MNPs on terrestrial mammals and humans, and to identify remaining critical knowledge gaps that could guide new scientific research directions. In this review, we summarize the important findings from the latest MNP toxicity studies in in vitro mammalian cellular and in vivo rodent models for each route of exposure. In addition, we discuss the findings of several recently published human biomonitoring studies, which have identified exposures of infants, children, and adults to MNPs, and characterized the absorption, bioaccumulation, and excretion of MNPs in the human body. We also review the impacts of MNP physico-chemical properties, including particle size, shape, surface charge, additives leaching and/or EP adsorption on their toxicity, and the potential cellular and molecular mechanisms that lead to toxicity upon MNP internalization. Finally, we identify and discuss the remaining critical knowledge gaps that must be bridged to complete our understanding of the potential hazards of MNP exposures and put forward recommendations for future research directions. The information presented in this review will help to improve our current understanding of the potential risks of MNP exposure in humans.

2. Methods

A combination of the terms “microplastic”, “nanoplastic”, “micro-nanoplastics”, “MNPs” and “toxicity” were used in the PubMed, Web of Science, and Google Scholar search engines to search for and identify scientific publications on MNP toxicity. The Google search engine and references sections of published papers found in the initial search were also used to identify additional articles related to the toxicity of MNPs. The search was restricted to peer-reviewed literature published in the English language from January 2015 through September 2022. The search results were limited to toxicity studies of exposures to micro- and/or nano-sized plastic particles across biological systems, including in vitro human and other mammalian cellular models and in vivo rodent animal models, as well as human MNP biomonitoring studies. Studies related to aquatic organisms and terrestrial plants were excluded from this review.

3. Results

A total of 220 peer-reviewed papers specifically related to MNP toxicity were found to meet the above criteria, with the annual number of publications growing exponentially over the last several years (Figure 1). These included 102 studies in in vitro cellular models, 100 studies in rodent animal models, 11 studies that included both in vitro cellular and in vivo rodent models, and 7 human biomonitoring studies, as shown in Figure 1. The different types of MNP particles employed in MNP toxicity studies are summarized in Figure 2. Each paper was thoroughly reviewed and categorized based on study model(s), employed MNP types, toxicological endpoints, and key results were extracted from each, as detailed in Table 2 and Table 3. For publications that used both in vitro and in vivo models, each component was represented separately.

Figure 1.

A. The number of peer-reviewed publications of MNP toxicity studies from January 2015 to September 2022. B. Types of MNP toxicity studies.

Figure 2. Types of MNP particles used in toxicity studies.

PS: polystyrene, PE: polyethylene, LDPE: low density polyethylene, PP: polypropylene, PVC: polyvinyl chloride, PET: polyethylene terephthalate, PA: polyamide, PU: polyurethanes, TW: tire wear.

Table 2. Summary of MNP toxicity studies in in vitro cell models.

PS: polystyrene, PE: polyethylene, PVC: polyvinyl chloride, PET: polyethylene terephthalate, PP: polypropylene, PA: polyamide, PU: polyurethanes, HDPE: high density polyethylene, LDPE: low density polyethylene, FS: face scrubs, TW: tire wear.

Cell lines	MNP Types	Endpoints	Notes	References
Gastrointes tinal cell models
Caco-2, HT29-MTX, AGS, GES-1, SNU-1, STC-1, KATO III, CCD-18Co	PS, PE, PVC, PET, PP	Particle internalization and translocation, cytotoxicity, viability, proliferation, lysosomal activity, apoptosis, autophagy, senescence, cell cycle, membrane integrity, morphology, oxidative stress, NO production, mitochondrial function, tight junction, inflammation, genotoxicity, cytoskeleton integrity, metabolism	Caco-2 cell line is the most common cell type for MNP toxicity studies	[38, 41, 65, 67, 68, 70-72, 81, 87, 88, 94, 97, 102, 103, 111, 112, 135, 137, 138, 219, 236] [50, 73, 163]
Caco-2/HT29-MTX or Caco-2/HT29-MTX/Raji B coculture cell models	PS, PE	Particle internalization, viability, cytotoxicity, oxidative stress, inflammation, intestinal barrier integrity, genotoxicity	Environmentally relevant MNPs; simulated digestion	[46, 136, 142]
Caco-2/HT29-MTX/THP-1 coculture cell models	PE, PVC	Viability, cytotoxicity, inflammation, intestinal barrier integrity, genotoxicity	Pathological model	[85, 93]
Caco-2/HT29-MTX/MDM/MDDC coculture cell models	PA, PU, PP	Cytotoxicity, viability, inflammation, membrane integrity, tight junction	Environmentally relevant MNPs	[51]
Respiratory cell models
BEAS-2B, A549, HPAEpiC, HNEpCs	PS, PET, PE	Particle internalization, viability, cytotoxicity, morphology, epithelial barrier integrity, oxidative stress, mitochondrial function, ER stress, inflammation, cell cycle, apoptosis, autophagy, NO production, epithelial-to-mesenchymal transition, metabolism		[75, 95, 98, 137, 143-148, 198, 237-240]
Other cell models
Liver: L02, HepG2, BNL CL.2, primary rat hepatocytes, HL7702, liver organoids	PS, PE	Particle internalization, viability, cytotoxicity, apoptosis, oxidative stress, mitochondrial function, inflammation, genotoxicity, activation of signaling pathways, lipid metabolism, metabolomics	Liver organ oid model; Environmentally relevant concentrations	[81, 153, 154, 163, 216, 217, 237, 241]
Kidney: HEK293, HK-2, MDCK	PS, PE	Particle internalization, viability, cytotoxicity, oxidative stress, ER stress, mitochondrial function, inflammation, necrosis, apoptosis, autophagy, metabolism, genotoxicity, kidney barrier integrity	Environmentally relevant concentrations	[155, 156, 242]
Placenta: HTR8/SVne o trophoblast, JEG-3, BeWo b30/HPEC-A2 coculture	PS	Particle internalization, cytotoxicity, viability, oxidative stress, inflammation, cell cycle, apoptosis, migration and invasion, genotoxicity	Determination of placental absorption and translocation	[243-245]
Embryo: Mouse embryonic fibroblasts (MEFs)	PS	Particle internalization, viability, oxidative stress, inflammation, genotoxicity, autophagy, apoptosis, morphology, transformation, invasion and migration	Long term coexpo sure to MNPs and arsenics	[246-248]
Skin: Human dermal fibroblasts (HDFs), HaCaT, KeraSkin™ 3D human skin culture model	PS, PE, PP, HDPE, LDPE, FS	Particle internalization, cytotoxicity, viability, oxidative stress, genotoxicity	Environmentally relevant MNPs skin irritation toxi city Study	[48, 49, 52, 57, 137, 233, 240, 249, 250]
Eyes: HCECs, HConjECs, MCTT HCE™ 3D human cornea culture model	PS, PP	Particle internalization, cytotoxicity, oxidative stress, proliferation, inflammation, apoptosis, eye irritation	Environmentally relevant MNPs eye irritation toxi city study	[57, 149]
Reproductiv e system: GC-2, HeLa, MDA-MB 231, MCF-7	PS, PE, HDP E, LDPE	Viability, proliferation, oxidative stress, ATP production, mitochondrial function, apoptosis, autophagy	Environmentally relevant MNPs	[43, 49, 50, 107, 214, 251]
Immune system: Raji-B, THP-1, RAW264.7, Jurkat, RBL-2H3, murine splenic lymphocyte s, U937, ImKC, J774A.1, HMC-1, TK-6, PBMCs, RBCs, hBM-MSCs	PS, PE, PET, HDP E, LDPE, PP, TW	Particle internalization, membrane integrity, viability, cytotoxicity, lysosomal activity, oxidative stress, NO production, inflammation, mitochondrial function, ATP production, autophagy, necrosis, necroptosis, macrophage polarization, cell cycle, differentiation, genotoxicity, lipid accumulation, macrophage metabolism and clearance, phagocytosis, hemolysis, histamine secretion, immune activation	Environmentally relevant MNPs	[42, 48-50, 136, 137, 161-163, 180, 205, 240, 252-259]
Vascular system: human umbilical vein endothelial cells (HUVECs), MyEND	PS	Particle internalization, membrane integrity, viability, cytotoxicity, oxidative stress, inflammation, mitochondrial function, autophagy, necrosis, cell adhesion, angiogenic tube formation, wound healing	Dynamic micro fluid exposure to particles	[150-152, 180, 260, 261]
Neural system: NSCs, T98G, SH-SY5Y, HMC-3, NE-4C, HT22, BV2, NS20Y, primary brain cells, primary brain vascular endothelial cells, primary mixed neuronal cells, primary astrocytes	PS, PE	Particle internalization, viability, cytotoxicity, ATP production, oxidative stress, cell cycle, phagocytosis, apoptosis, microglial activation, immune responses, cell metabolism, lipid accumulation, reactive astrocytosis, expressions of neuronal and glial markers, activation of signaling pathways	Dynamic microfluid exposure to particles in cells Environmentally relevant	[62, 159, 160, 162, 248, 251, 262-264]

Table 3. Summary of MNP toxicity studies in in vivo rodent models via different exposure routes.

PS: polystyrene, PE: polyethylene, PET: polyethylene terephthalate, PVC: polyvinyl chloride, PA: polyamide, PP: polypropylene, TW: tire wear.

Exposure routes	Animal models	MNP types	Endpoints	Notes	References
Ingestion	Mice, rats	PS, PE, PET, PVC, PA, PP	Particle biodistribution and bioaccumulation, oxidative stress, genotoxicity, histopathological changes, pathological changes, serum biochemistry, gut barrier integrity, microbiota homeostasis, energy metabolism, locomotor activity, behavioral activity, reproductive function, developmental function, AChE activity, muscle function	Environmentally relevant MNPs; Co-exposure to MNPs and other chemicals	[187, 198, 261]
Inhalation	Mice, rats	PS, TW	Particle biodistribution and bioaccumulation, pulmonary function, bronchoalveolar lavage fluid (BALF) analysis, oxidative stress, pathological changes, AChE activity, hematological changes, serum biochemistry, locomotor activity, behavioral activity, reproductive function	Environmentally relevant MNPs	[34, 35, 147, 198-200, 265]
IP injection	Mice	PS	Particle bioaccumulation, oxidative stress, genotoxicity, AChE activity, serum biochemistry, brain biochemistry, behavioral activity, reproductive function	Co-exposure to MNP sand Zincoxide	[106, 201]
IV injection	Mice	PS	Oxidative stress, histopathological changes, oxidative stress, hepatic lipid metabolism, reproductive function, developmental function	High-fat diet fed mouse model	[266, 267]
Eye contact	Mice	PS	Eye structure, tear secretion, inflammation of the lacrimal gland and conjunctiva		[149]

3.1. MNP materials used in MNP toxicity studies

Most MNP toxicity studies to date have been carried out using commercially available primary MNPs of polystyrene (PS), polyethylene (PE), polyvinyl chloride (PVC), polyethylene terephthalate (PET) and other polymers. These are typically monosized symmetrical or spherical particles whose surface chemistries are those of the corresponding pristine polymers. However, the majority of MNPs that contaminate the air, water, and food and thus the MNPs to which humans are most likely to be exposed, are secondary MNPs, generated through degradation and fragmentation of plastic materials. By contrast to primary MNPs, such secondary MNPs tend to have irregular shapes, varied sizes, and mixed compositions [45]. Some recent studies have investigated the toxicity of such environmentally relevant secondary MNPs. The various types of secondary MNPs employed in these studies, produced under laboratory conditions from pristine plastics by either thermal decomposition, mechanical fragmentation, photoaging, or a combination of these, are summarized in Table 1.

Table 1. Summary of environmentally relevant secondary MNPs used in MNP toxicity studies.

Materials	Generation methods	References
PE	Incineration, milling, grinding	[46, 47, 52, 54]
PS	Milling, photoaging	[50, 58, 112]
PP	Milling, grinding, photoaging	[48, 51, 53, 54, 57]
LDPE	Milling	[49]
PU	Milling, photoaging	[51, 56]
TW	Photoaging	[55]
PET	Grinding	[53]
PA	Milling	[51]

Incineration has been employed to achieve the thermal decomposition of bulk-size virgin PE pellets, producing incinerated PE (PEI) MNPs [46, 47]. Mechanical fragmentation of large plastics under laboratory conditions has included cutting, milling, grinding, and sieving. Milling and grinding processes fragment plastics into small particles, which are considered simple and effective ways to mimic the mechanical stressors occurs in natural environments. Environmentally relevant secondary polypropylene (PP) MNPs are generated from ball milling of solid PP particles to mimic the physical abrasion that occurs in the natural weathering process [48]. Likewise, ball milling is used to produce environmentally relevant low-density PE (LDPE) [49] and PS MNPs [50]. Grinding has also been used to produce secondary MNPs. Polyamide (PA), polyurethane (PU), and PP plastics are fragmented using an impeller breaker to produce MNPs with mean diameters of 300 μm [51]. Nano-sized PE MNPs are generated by grain grinding of PE pellets and subsequent homogenization and ultrasonication [52]. Plastic pieces cut from PP and PET food containers, regarded as real-life plastic products, were ground using blenders and subsequently filtered and desiccated to generate nano-sized MNPs for use in toxicity studies [53]. Micro-sized MNPs collected from plastic waste, which consisted primarily of PE and PP particles with heterogeneous shapes and sizes, were further mechanically ground and sieved to generate MNP granules with sizes of 50 -100 μm and <50 μm [54]. Photoaging driven by sunlight radiation is another environmental stressor that can degrade plastic products into low molecular weight polymer fragments through photo-initiated oxidative reactions [9]. During photoaging, persistent free radicals and dissolved organic matter released from MNPs may pose potential adverse effects on human health [55, 56]. A more comprehensive multistep weathering process employing cryomilling, sieving, heating, and UV radiation to simulate both mechanical abrasion and photoaging in the environment was used by Kim et al. and Volkl et al. to produce weathered secondary PP and PS MNPs [57, 58]. Such environmentally relevant secondary MNPs, generated from natural or simulated environmental degradation of plastics, have irregular and varied shapes, sizes, and surface topologies, as well as altered surface chemistries, including bio-reactive oxygen-containing surface functional groups. Their physicochemical characteristics thus differ substantially from those of the commercially available, uniform and pristine, primary MNPs most commonly used in MNP toxicity studies to date. The use of environmentally relevant secondary MNPs in biological studies is expected to provide more meaningful toxicological data and an accurate evaluation of the biological impacts of real-world MNPs on humans and other organisms.

3.2. The role of physico-chemical properties in MNP toxicity

The toxicity of MNPs is largely dependent on their physico-chemical characteristics, including size, shape, chemical composition, porosity, specific surface area, surface chemistry and charge, surface roughness, and others. Since MNPs vary widely in such physico-chemical characteristics, the results of toxicity testing using different MNPs are very diverse. Moreover, the toxicological findings for a given MNP may vary considerably among experimental test models (animal species, age, and sex, or in vitro model cell type and format), as well as between different exposure scenarios. Unfortunately, the most consistent feature among the published toxicological studies of MNPs to date is a lack of standardization.

3.2.1. Size

Size is considered the most critical factor affecting the absorption, bioaccumulation, biodistribution, and thus in large part the toxic effects of MNPs. Large MNP particles are not absorbed by the gut epithelium and instead remain within the mucus layer, which leads to local inflammation and immune responses [59]. In contrast, smaller MNP particles (<150 μm) can more easily cross the gastrointestinal epithelium and enter the circulation, leading to widespread systemic exposure in mammals [4, 60]. The bioaccumulation and biodistribution of MNP particles within mammalian bodies are highly dependent on particle size. Deng et al. found that 20 μm microplastic particles can enter and accumulate in mice’s gut, liver, and kidney [37]. However, the bioaccumulation data in this study was questionable as discussed in Braeuning’s report [61]. Other studies should thus be considered to discuss the bioaccumulation and biodistribution of MNPs. Somewhat smaller micro-scale MNPs (≤ 10 μm) can reach a greater number of organs and penetrate biological barriers, including the blood-placenta barrier, blood-testis barrier, and blood-brain barrier [36, 62-64]. Nanoscale MNPs can easily translocate and deposit in a large number of organs. 50 nm PS MNPs were found in the walls of the GI tract, kidney, and heart after single-dose oral exposure in rats [39], and 60 nm PS MNPs were found in the stomach, liver, and kidney after single-dose oral exposure in mice [38]. Subacute (28 d) oral exposure of 100 nm carboxylated PS MNPs in mice led to the accumulation of particles in the GI tract, spleen, lung, kidney, testis, and brain [65]. Moreover, smaller MNPs resulted in greater tissue accumulation and higher biokinetic constants in mice [66]. In a notable exception, 4 and 10 μm PS MNPs were found to accumulate within mouse testis than 0.5 μm PS MNPs 24 h after MNP exposure [64].

It is worth noting that the size-dependency of toxic effects caused by MNP exposure in animal models is controversial. For example, large-sized PS MNPs caused higher mitochondrial depolarization compared to small-sized particles [67, 68]. Meng et al. found that the aggregation of 600 nm PS MNPs in the kidney enhanced their toxicity relative to both larger and smaller non-aggregated MNPs [69]. However, at the cellular level, the uptake and cytotoxicity of MNPs are greatly influenced by particle sizes and exhibit strong size dependency. Nanosized MNPs had a higher rate of internalization and accumulation in cells compared to microscale MNPs [34, 46, 70]. Smaller-sized PS MNPs also led to greater cytotoxicity, including loss of cell viability, oxidative stress, inflammation, mitochondrial dysfunction, DNA damage, and cell apoptosis [41, 71, 72]. Bonanomi et al. found that 0.5 μm PS MNPs were more easily internalized compared to 2 μm after 48 h exposures, although after extended exposure the internalization of 2 μm PS MNPs increased and after 4 weeks was comparable to that seen with the smaller particles at 48 h [73]. Moreover, 0.5 μm MNPs were more easily removed from cells after 4-week exposure compared to 2 μm particles. Recent findings have also raised concerns about the interaction of small-size MNPs, especially nanosized particles, which can enter cell organelles, including nuclei, mitochondria and lysosomes, potentially leading to further toxic effects, such as DNA damage, metabolic dysfunction, and cell death [46, 74, 75]. It is worth noting that surface area, surface-to-volume ratio, and surface reactivity can vary greatly with particle size, which may further modulate MNP bioactivity.

3.2.2. Shape

Shape is another physical characteristic that decisively affects MNP toxicity by altering the interactions of MNPs with cells and tissues. MNP particles exhibit a wide variety of shapes, including beads, fibers, fragments, films, sheets, filaments and foams [76, 77]. Commercial micro- and nanosphere plastics can be easily obtained and are commonly used for toxicity studies. These uniformly shaped nano- and microspheres are useful tools for establishing the basic models of risk assessment for MNP exposure. However, exposures to such uniform and regularly shaped particles may not exhibit the same biological interactions and responses in experimental systems as the environmentally relevant real-world MNPs to which humans are exposed, which comprise a heterogenous mixture of irregularly shaped particles and fibers. Microfibers are a major source of microplastics in the environment [78]. The thin and elongated shape of microfibers allows them to easily penetrate the lung and avoid engulfment by macrophages. Chen et al. reported finding 24 types of microplastics in microfiber form in human lung tissue [79]. More toxicity studies employing environmentally and human exposure-relevant MNPs with irregular shapes are urgently needed.

3.2.3. Chemical composition/polymer type

Although the toxicity of MNPs greatly depends on their size and shape, the effects of polymer types should not be disregarded. The most common polymers detected in freshwater and drinking water are non-biodegradable PE, PP, PS, PVC and PET [80]. Since each different plastic polymer clearly has a different and unique chemical structure, it would be reasonable to expect that each would have different and possibly unique interactions with cells and biomolecules, and thus potentially different biological effects. In a comparison of in vitro uptake and toxicity of MNPs of varied chemical composition, including PE, PP, PET, and PVC, using Caco-2, HepG2, and HepaRG cell models, Stock et al. showed that uptake and toxicity were both size- and material-dependent [81]. To date, however, studies of PS have dominated the research literature on MNP toxicity and biointeractions. This discrepancy is due to the fact that PS microspheres and nanospheres are widely commercially available in a broad range of sizes, with a variety of surface modifications, and with or without a variety of embedded fluorophores, whereas there is an almost complete lack of commercial MNP particles of other polymers. While PS is certainly one of the major plastic polymers produced and found in environmental samples, its production and occurrence in the environment are dwarfed by those of PE polymer, which is the most common polymer type found in plastic products and environments [82-84]. However, limited in vitro data are available on the toxicity of PE MNPs. An additional obstacle to studies of some MNP polymers is their low density and thus buoyancy in cell culture media, requiring adaptation of cell culture methods to provide contact of these MNPs with cells. An inverted transwell system was employed by Stock et al. to study the toxicity of buoyant PE and PP MNPs [81], and Busch et al. applied an inverted triple cell culture model to identify the cytotoxicity, inflammation, and intestinal barrier dysfunction caused by PE MNP exposure [85]. PE MNP exposure via oral administration also caused microbiota dysbiosis and gut inflammation in mice [86]. Likewise, toxicity studies employing MNPs of PET, the most highly produced of all plastic polymers (think water bottles) are very limited. However, in those studies that have employed PET MNPs so far, no significant toxic effects were observed [87, 88].

Additives, including plasticizers, pigments, stabilizers, flame retardants, UV stabilizers and antioxidants, are added routinely during the polymer manufacturing process to improve the plastic processibility and stability and to prolong the plastic’s lifespan. These additives can leach into the air, water, foods, or even human blood and other tissues from MNPs under different environmentally or biologically relevant conditions, posing risks to ecosystems and human health [89, 90]. Deng et al. found that MNPs could transport and release plastic additive phthalate esters (PAEs) into the guts of mice and that PAE-contaminated MNPs can cause significant gut dysfunction [91]. Further studies showed that PAE-contaminated MNPs released PAEs in the mouse gut, liver, and to a lesser extent in the testis, and PAE-contaminated MNPs were found to increase male reproductive toxicity compared to exposure to MNPs or PAEs alone in mice [92].

3.2.4. Surface properties

The effects of MNPs on cells and tissues are also dependent on their surface properties, including functionalization, charge, hydrophobicity, surface area, and roughness. These characteristics influence the dispersion and aggregation of the particles. Moreover, with changes in surface properties, the interactions between MNP particles and cells could be altered, potentially affecting the particle interactions at the cell surface, internalization by cells, and subsequent intracellular interactions, bioactivities, and toxicities. MNP surfaces and their charges can be modified by grafting differently charged functional groups. The two most common functional modifications of the MNP surface are amination and carboxylation, which affix positively charged (at physiological pH) amine and negatively charged (at physiological pH) carboxyl groups, respectively, to the MNP surfaces. It should be mentioned that such modifications are only commercially available for PS micro- and nanospheres, and thus all studies of the effects of surface functionalization or charge to date have therefore employed PS MNPs. In studies of such functionalized PS MNPs, the presence of that such functionalization has been shown to result in greater cellular uptake and toxicity relative to non-modified PS MNPs [46, 65, 70, 93, 94]. In addition, positively charged amine-modified particles were found to be more easily internalized by cells and to induce greater cytotoxicity than their negatively charged carboxylated counterparts [70, 95]. Amine-modified PS MNPs have also been shown to have strong interactions with mucin and to induce mucin aggregation independent of cell viability or apoptosis, while also decreasing cell viability and increasing apoptosis to a significantly greater extent than either unmodified or carboxylated PS MNPs in both mucin-secreting and non-mucin-secreting cells [96]. In contrast, Walczak et al. demonstrated that carboxyl-modified PS MNPs were more readily taken up and had a wider biodistribution than neutral or amine-modified PS MNPs in rats [39].

Surface area is correlated with particle size, surface roughness, and surface functionalization. Qin et al. showed that treatment of PS MNPs with chlorine (simulating disinfection of drinking water) resulted in the formation of carbon-chlorine bonds and free radicals on the MNP surfaces and significantly increased surface roughness; and that the resulting chlorinated PS MNPs caused significantly greater cytotoxicity in human GES-1 cells than pristine PS MNPs, which the authors attributed to the enhanced surface roughness and altered surface chemistry of the chlorinated PS MNPs [97] Moreover, coating of PS MNPs with a protein corona, as would likely occur in any human exposure before MNPs came into contact with cells, and which would dramatically alter the effective surface chemistry of an MNP, was found to significantly reduce cytotoxicity in A549 cells, with a soft coronae reducing cytotoxicity to a greater extent than a hard corona [98]. It has also been shown that MNPs with greater surface area are able to adsorb larger amounts of various hazardous contaminants, including heavy metals (cadmium, mercury, arsenic, etc.), antibiotics (amoxicillin, tetracycline, etc.), endocrine disrupters (polycyclic aromatic hydrocarbons, polychlorinated biphenyls, etc.), microorganisms, and other inorganic and organic chemicals from aquatic environments [24, 71]. The potential bioavailability of these contaminants associated with MNP particles will depend on their physical and chemical interactions with the particle body [99]. MNPs could serve as transportation vectors for such pollutants into organisms and cells, therefore presenting additional risks to human health [100]. A number of studies have revealed synergistic or additive toxicity caused by co-exposure to particles with contaminants, including increased cellular internalization [87], exacerbated oxidative stress, disrupted ATP-binding cassette (ABC) transporter and cytoskeleton [41], neurotoxicity, metabolic toxicity [101], gut dysfunction and inflammation [91], microbiota dysbiosis [102], aggravated reproductive toxicities [92], and genotoxicity [103]. Some studies have suggested that MNPs could also serve as sinks for some contaminants in the marine environment [104]. However, no clear data have indicated the formation of stable MNP-chemical aggregates resulting in further deposition within the body. Co-exposure of mice to PS MNPs and tributyltin (TBT) ameliorated the toxic effects on the gut and liver caused by exposure to PS MNPs or TBT alone [105]. Similarly, combined exposure to ZnO and PS MNPs ameliorated neurotoxicity and cognitive impairment caused by treatment with either material alone, which the authors attributed to aggregation of the MNPs in the presence of ZnO, resulting in decreased translocation of the larger-sized agglomerates into neural tissues [106]. The stable binding of some substances on MNPs may enable their transport to unexpected areas and/or sink within the body, resulting in lower toxicity.

Chemical additives released during the degradation of plastic debris can also be re-adsorbed onto MNPs, further exerting indirect effects on human health. For example, the adsorption of bisphenol A onto nanoscale PS MNPs exacerbated the toxicity caused by the MNPs alone in Caco-2 cells [68]. Studies have also found that tri-o-cresyl phosphate, a plasticizer found in plastic materials, and parabens, an endocrine-disrupting hormone that adsorbs to MNPs, interact with estrogen receptors on breast cancer cells to promote tumor growth [107, 108]. Likewise, co-exposure to tetrabromobisphenol A (TBBPA) and PE MNPs exacerbated cytotoxicity in Caco-2 cells and microbiota dysbiosis [102]. More studies are needed to further examine the role of MNPs in the toxicity of co-contaminants.

MNP particles can interact with various biomolecules in physiological systems and environmental elements to alter their physico-chemical properties, including structural changes, protein corona formation, and increased surface area [69]. Biotransformation of ingested MNPs during digestion can have important impacts on MNP biointeractions in the gut. It has been shown that the interaction of PS MNPs with simulated digestion fluids causes physical-chemical transformation of the MNPs, resulting in their agglomeration and accompanied by adverse effects, including disruption of intestinal epithelial barrier integrity, cytotoxicity, and cell death [46, 69]. Liao et al. employed an in vitro full digestive system simulation, including oral, gastric, small intestinal, and large intestinal phases, to investigate the desorption of Cr by Cr-loaded MNPs of several polymers across the GI tract. They found that desorption, and thus bioaccessibility of Cr, was significantly increased in the gastric phase compared to other phases of digestion [109]. Bakir et al. found that desorption of several persistent organic pollutants (POPs) from PE and PVC MNPs was up to 30 times greater under simulated gut conditions than in seawater, thus increasing the potential for POP toxicity [110]. In contrast, the toxicity of PS MNPs in Caco-2 cells was mitigated by the formation of a surface corona on the MNPs during simulated digestion [111].

MNPs undergo photo-oxidative and hydrolytic reactions (aging/weathering) in the environment, resulting in the formation of surface functional groups such as carboxyl and carbonyl, depending on the polymer [9]. Such transformations induced by UV radiation were shown to increase MNP surface roughness and modify surface functional groups of MNPs, leading to increased toxicity in Caco-2 cells, including disruption of cell membrane integrity and loss of cell viability [112].

3.3. MNP toxicity studies

The general goal of MNP toxicity studies is to determine the dose-response relationship between MNP exposure levels and potential adverse effects across the three major routes of exposure, namely ingestion, inhalation, and dermal contact. Because MNPs comprise a highly diverse class of materials, a comprehensive and systematic approach is required to organize the available data on the toxicological effects of MNPs on cells, tissues, organs, and the body, and to further identify critical knowledge gaps and directions for future research. We included findings from recent human biomonitoring studies to evaluate human exposure to MNPs and potential health risks. We also reviewed MNP toxicity studies performed in in vitro experiments using cell lines and in vivo experiments using laboratory animals.

3.3.1. Human biomonitoring studies

MNPs may cause potential hazards to humans via different exposure pathways, including ingestion, inhalation exposure from airborne MNPs, or dermal contact. A variety of MNP factors, including particle polymer type, size, shape, and surface chemistry, as well as gender differences and geographical and socioeconomic environmental factors, can affect human MNP exposures [113, 114]. Food and beverage consumption is the primary source of MNP ingestion exposure in humans. MNP particles are found in many foods and drinks, and are also released into foods and beverages from plastic packaging [80, 115-118]. Contamination of meat and fish with MNPs can also occur through trophic transfer in natural and farmed aquatic and terrestrial food chains [119, 120]. One study has estimated that the average human ingests 5 grams of plastic (the equivalent weight of one credit card) per week [121].

Airborne environmental and occupational MNP particles are the primary sources of MNP inhalation exposure in humans. Sources for these airborne MNPs include microfibers from synthetic textiles, carpets and furnishings, microplastics from tire abrasion, and MNPs released from printing equipment and incineration facilities. Vianello et al. estimated that human inhalation exposure to microplastics is up to 272 particles per day in a breathing thermal manikin model [122]. Huang et al. identified various types of micro-sized MNPs in sputum from each of the 22 human subjects, among them PU was dominant [123]. Airborne fine particles (PM_2.5, particle size < 2.5 μm) usually deposit in the respiratory tract and are further removed by mucociliary clearance, whereas ultrafine particles (particle size < 100 nm) can reach the pulmonary alveoli and translocate through the blood-air barrier to the circulatory system, suggesting the possible distribution of nanosized MNPs in the human body [114, 124].

Dermal contact is considered a more unusual exposure route, the sources of which are personal care products, dusts, and fabrics. Skin can act as a highly protective barrier, with only nanoparticles having sizes less than 4 nm able to penetrate intact skin, although damaged skin can allow penetration of particles up to 45 nm in size [125]. There are no current studies assessing human dermal exposures to MNPs and associated potential risks.

The extensive use of PPE during the COVID-19 pandemic period has attracted increasing attention to the release of and exposure to MNPs. It has been reported that vast amounts of MNPs were likely released from PPE into the aquatic environments [126], with more than 1.5 million micro-sized MNPs released from a single weathered mask [127]. Li et al. found that the release of MNPs from commonly used masks is greatly affected by the mask type and disinfection pretreatment processes employed [128].

The potential risks of MNPs to humans are receiving increasing attention since MNPs have been found in a wide range of human samples. Schwabl et al. identified 9 common types of microplastics in human stool samples in a prospective case series [129]. MNPs of several major plastic polymers have been identified in human blood samples from volunteers (at an average total of 1.6 μg/mL), a clear indication that MNPs, inhaled, ingested, or otherwise, can reach the human bloodstream and potentially be transported throughout the entire body [130]. MNPs of varied polymers, sizes, and shapes have been identified in both fetal and maternal zones of the human placenta as well as in fetal meconium, strongly suggesting maternal-fetal transfer of MNPs and a significant additional potential risk to successive generations of humans [131, 132]. In addition, MNPs have been identified in human breast milk, further suggesting widespread translocation in humans and potentially posing a great threat to infants [133]. However, there are very limited data specifically addressing human MNP exposure, biodistribution, and pathophysiological/toxicological effects. In part this is due to a lack of accurate methods for quantifying MNPs, particularly those in the nanoscale size range, in biological samples. Yang et al. developed a toxicity-based toxicokinetic/toxicodynamic (TBTK/TD) model to determine TK parameters and dose-response profiles in mice, thus predicting human exposure threshold and implicating potential human health risks [66]. Further epidemiological research is urgently needed to address and evaluate the potential risk of MNPs to human health.

3.3.2. In vitro MNP toxicity studies

A variety of in vitro cellular models employing well-characterized primary cells or immortalized cell lines derived from human or animal tissues have been used to assess the toxicity and bioactivities of MNPs for each major route of exposure. These studies are summarized in Table 2.

3.3.2.1. Gastrointestinal cell models (ingestion route of exposure)

Ingestion is considered one of the most dominant routes of MNP exposure in humans. Cellular models based on immortalized gastrointestinal cell lines are considered the most suitable in vitro platforms for studying the potential toxicity of such exposure to ingested MNP. The most commonly used small intestinal cellular model is a monoculture of the immortalized Caco-2 human colon adenocarcinoma cell line. After 2-3 weeks in culture, the cells differentiate into a monolayer that closely recapitulates the structure and function of small intestinal enterocytes [134]. In the Caco-2 monolayer model, PS MNPs produce size-, exposure time- and concentration-dependent particle accumulation and cytotoxicity [65, 103, 112, 135]. PS MNPs internalized by the differentiated Caco-2 cells [136] decreased cell viability [67], increased reactive oxygen species (ROS) production [135], impaired intestinal barrier function, disrupted cell junctions, and altered expression of genes related to stress and inflammation [103]. However, no significant cell damage was observed after 24 h exposure of Caco-2 cells to PE MNPs (~30.5 and 6.2 μm) at high concentrations of up to 1000 μg/mL [137]. In the colorectal adenocarcinoma cell line, HT29-MTX, which phenotypically resembles mucus-secreting small intestinal goblet cells, acute (24 h) exposure to 3 and 10 μm PS MNPs caused moderate cytotoxicity, lysosomal membrane permeabilization (more pronounced with 3 μm PS MNPs), and increased ROS production (more pronounced with 10 μm PS MNPs), which was partially reversed after subchronic (48 d) exposure [138]. Several different gastric cell models have also been used to assess ingested MNP toxicity. In human gastric adenocarcinoma (AGS) cells, internalized nanosized PS MNPs decreased cell viability, altered cell morphology, induced inflammation, oxidative stress, and cell apoptosis after 24 h exposure in a concentration-dependent manner [71, 72]. Using the Comet assay, these studies also provided evidence that 60 nm PS MNP exposures induced DNA damage. In human gastric epithelial GES-1 cells, 60 nm PS MNPs were internalized and caused significant cytotoxicity [38]. In human gastric epithelial SNU-1 cells, PS MNPs caused toxicity that was dependent on particle size, surface modification, and exposure time [70]. The colonic cellular model human colon fibroblast CCD-18Co cells, acutely (4 h) or chronically (4 weeks) exposed to 0.5 and 2 μm PS MNPs, showed oxidative stress and metabolic rearrangement but no evidence of cytotoxicity [73]. These observations demonstrate that the cellular internalization and cytotoxic effects of MNPs in gastrointestinal cells are dependent on particle sizes and surface modifications, and exposure concentrations and durations.

Although studies in monoculture models such as those described above can provide useful toxicological data, the in vivo gastrointestinal epithelium is far more complex and cannot be modeled as a single cell type. Advanced cellular models that more accurately recapitulate the organization, structure, and function of the complete intestinal epithelium are required to provide more meaningful and predictive results. A more complete and physiologically relevant model of small intestinal epithelium can be created by coculturing Caco-2 and HT29-MTX cell lines. Although both lines are derived from colon adenocarcinoma cells, after about 2 weeks of coculturing constant culture, the Caco-2 cells acquire a small intestinal enterocyte phenotype, while HT29-MTX cells acquire a mucus-secreting goblet cell phenotype. The co-differentiation results in a mixed monolayer including absorptive enterocytes, which typically constitute about 80% of cells in the small intestinal epithelium, and mucus-secreting goblet cells, which constitute approximately 4%, 6%, and 12% of intestinal epithelial cells in the duodenum, jejunum, and ileum, respectively [139]. The differentiated HT29-MTX cells provide an apical mucus layer similar to that of the small intestine. The further addition of Raji-B (feeder) cells to the Caco-2/HT29-MTX coculture media after initial differentiation induces a small portion of Caco-2 enterocytes to further differentiate into cells resembling intestinal epithelial microfold- or M-cells. These cells translocate bulk intestinal luminal fluid to submucosal lymphocytes of the gut-associated lymphoid tissue (GALT) to provide antigenic surveillance of the intestinal contents [140, 141].

To date, only a few studies have employed such models for in vitro assessment of the biological effects of MNPs in the small intestine [46, 47, 136, 142]. Stock et al. found that both 1 and 4 μm PS MNPs were internalized to a significantly greater extent in the Caco-2/HT29-MTX coculture than in the Caco-2 monoculture after 24 h exposure [136], which is somewhat surprising given that the mucus layer in the coculture model would be expected to impede access of the MNPs to the enterocyte surface. Domenech et al. reported concentration-dependent internalization of nanosized PS MNPs after 24 h exposure, but no associated intestinal barrier dysfunction, cytotoxicity, or genotoxicity in either Caco-2/HT29-MTX coculture or Caco-2/HT-29-MTX/Raji B triculture models [142]. In contrast, DeLoid et al. observed size-dependent uptake of carboxylated PS MNPs accompanied by significantly reduced viability and increased epithelial barrier permeability at concentrations of 0.4 and 1 mg/mL in a Caco-2/HT29-MTX/Raji B triculture model [46]. Using a transwell triculture of Caco-2, HT29-MTX and THP-1 macrophages, Busch et al. observed that whereas neither PE nor PVC MNP exposures caused significant adverse toxic effects after 24 h, upon activation of pro-inflammatory pathways (by treatment with IFN-γ and LPS) exposure to PVC MNPs (but not PE MNPs) resulted in significantly increased IL-1β release and loss of epithelial cells [93]. In an inverted version of the Caco-2/HT29-MTX/THP-1 transwell model, the same investigators found that PE MNPs caused significant cytotoxicity and pro-inflammatory effects, which the authors attributed to the assumed buoyancy of PE MNPs [85]. In contrast, using a three-dimensional transwell model consisting of apical Caco-2 and HT29-MTX cells and basolateral human monocyte-derived macrophages (MDMs) and dendritic cells (MDDCs), Lehner et al. observed no significant cytotoxicity, effects on barrier integrity, or release of pro-inflammatory cytokines after exposure to environmentally relevant secondary PP, PU, and PA MNPs [51]. These studies indicate that the coculture or triculture gastrointestinal model provides a more complex and physiologically accurate model for MNP ingestion exposure and toxicity. The MNP toxicity in these models was size-, surface-modification, and concentration-dependent.

3.3.2.2. Respiratory cell models (inhalation route of exposure)

Since inhalation is also considered an important route of human exposure to MNPs, models based on respiratory cells are essential for in vitro assessment of MNP toxicity. Nanosized PS MNPs were found to accumulate within A549 cells (human lung carcinoma alveolar Type II epithelial cells) after 24 h exposure [98, 143]. Moreover, 20 nm PS MNPs were seen adjacent to chromosomes within A549 cell nuclei, indicating the potential for interactions of PS MNPs with cellular DNA. Exposure to MNPs caused concentration-, duration-, and cell type-dependent cytotoxicity, including decreased viability, increased oxidative stress, inflammatory responses, epithelial barrier dysfunction, altered cell cycle and ultimately cell death, in a variety of respiratory cell types, including A549 cells, normal human bronchial epithelial (BEAS-2B) cells, human pulmonary alveolar epithelial (HPAEpiC) cells, and human nasal epithelial cells (HNEpCs) [144-147]. For example, nanosized PS MNPs significantly decreased cell viability in A549 cells at high concentrations (25 nm PS at 25 and 30 μg/mL and 70 nm PS at 160, 220, and 300 μg/mL) after 24 h exposure [143]. However, no significant decrease in cell viability was seen in A549 cells exposed to 1 or 10 μm PS MNPs even at high concentrations (100 μg/mL) up to 72 h exposure, altered metabolic activity, inhibition of cell proliferation, and morphological changes were identified, suggesting potential epithelial injury [148]. Likewise, cytotoxicity combined with signs of altered metabolic activity, including increased concentrations of amino acids and tricarboxylic acid cycle intermediate metabolites, was observed in PS MNP-treated BEAS-2B cells at ≥ 10 mg/ml after 24 h, suggesting some resistance of these cells to frank toxicity after MNP exposure [75].

3.3.2.3. Other cell models (other routes of exposure and systemic exposure)

Studies investigating the potential impacts of dermal and ocular MNP exposures, and of systemic exposures following translocation of MNP particles from the primary site of exposure (gastrointestinal or respiratory exposure) have employed multiple cellular models, including dermal, corneal, vascular, hepatic, renal, neural, and immune cell lines, to assess MNP toxicity in vitro.

Dermal contact is considered a potentially important route of human MNP exposure. Nanosized MNPs isolated from commercial face scrubs and PE MNPs were internalized by human HaCaT keratinocytes, and induced time- and concentration-dependent ROS production as well as cytotoxic, cytostatic, and cytoprotective activity [52]. 50 nm or 2 μm PS MNPs were observed within the cytoplasm of human corneal epithelial cells (HCECs) and human conjunctival epithelial cells (HConjECs) with a higher amount for 50 nm after 48h exposure at 25 μg/mL, and caused oxidative stress and apoptosis in these cells [149]. However, 85.2 μm weathered PP MNPs with an irregular shape caused no skin or eye irritation in 3D human skin and corneal culture models after 3h exposure at the concentration of 66.7 mg/surface area of cultured cells per well, suggesting that these particles may not have the ability to penetrate multiple keratin or cellular layers [57].

Vascular cells are the primary target when small-sized MNPs are absorbed via the gut and reach the circulatory system. Lu et al. found that low interaction of 1 μm PS MNPs with human umbilical vein endothelial cells (HUVECs), but no significant toxic effects of these particles were identified even at the high concentration of 25 μg/mL after 48h exposure, whereas 100 and 500 nm PS MNPs interacted with these cells efficiently and induced cell membrane damage and autophagy at 25 μg/mL after 48h exposure [150, 151]. Lee et al. also identified that 48 h exposure of HUVECs to 500 nm MNPs caused autophagy- and necrosis-mediated cytotoxicity, and inhibited angiogenic tube formation, angiogenic signaling pathways and wound healing and cell migration events in a concentration-dependent manner, which can be a sign of endothelial dysfunction [152].

1 μm PS MNPs have been found to cause concentration-dependent hepatotoxicity and lipotoxicity in liver organoids by inducing oxidative stress and inflammatory responses and to impact ATP production and lipid metabolism by disrupting enzyme function [153]. PE MNPs with average sizes of 40–48 μm were found to cause a 171% increase in triglyceride content in human hepatoma (HepG2) cells after 48h exposure at the concentration of 50 mg/L [154]. However, neither HepG2 nor hepatic HepaRG cell lines showed signs of cytotoxicity when treated with environmentally relevant concentrations of various MNP particles [81].

Most pollutants that enter the body’s circulatory and lymphatic systems pass through the kidneys. Thus, the kidneys are likely to encounter a relatively large amount of MNPs that are taken up by the gut or translocated across the alveolar membrane in the lung to reach the circulation after ingestion or inhalation exposures, respectively. Upon exposure, microsized PS MNPs adhered to embryonic kidney 293 (HEK293) cells and human kidney proximal tubular epithelial (HK-2) cells and were internalized by these cells [155, 156]. Moreover, This exposure induced toxic effects, including reduced cell viability, oxidative stress, mitochondrial dysfunction, ER stress, inflammation, and autophagy. Cytotoxic concentrations (300 ng/mL) of PS MNPs can impair kidney barrier integrity, which could increase the risk of renal disease [156].

Nanoparticles can reach the brain parenchyma either by crossing the blood-brain barrier (BBB), which is comprised of tight junctions between cerebral microvascular cells and perivascular pericytes and astrocytes, or through the olfactory portal, which bypasses the BBB, providing direct access to the olfactory bulb and brain [157, 158]. Studies have shown that both pristine and carboxylated PS MNPs can accumulate in microglial cells, induce changes in cellular morphology and microglial immune activation, and further cause cell apoptosis [62, 159]. In neuroblastoma cells differentiated into neuronal types, nanosized PS MNPs induce cytotoxicity comparable to or even greater than that induced by the neurotoxin acrylamide. PS MNPs also caused shrinkage of neurite outgrowth, morphological alternations, nuclear swelling, and release of intracellular components [160]. After extended storage of MNP particles (shelf-life > 6 months), cellular uptake and toxicity of aged PS MNPs were found to be significantly increased compared to that of fresh PS MNPs, which could be attributed to particle aggregation and adsorption of bioactive compounds onto the surfaces of the particles [159]. Thus, it is important for in vitro studies of MNPs to utilize freshly prepared and well-characterized MNP particle suspensions to avoid differences in bioactivity resulting from unknown changes in MNP physicochemical and colloidal characteristics.

Several in vitro studies have investigated the effects of MNP exposure on cells of the innate and adaptive immune systems. Exposure of macrophages, lymphocytes, and lymphoblasts to MNPs resulted in oxidative stress and pro-inflammatory responses. The 24h exposure of Raji-B B lymphocytes to nanosized PS MNPs at the highest concentration decreased cell viability, increased ROS production and caused oxidative DNA damage [161]. In macrophage-like Raw 264.7 cells, nanosized pristine PS MNP exposure increased ROS production and impaired lysosome function [162]. Moreover, exposure to nanosized sulfate-modified (negatively charged) PS MNPs promoted the accumulation of lipid droplets in these macrophages and further induced the differentiation into lipid-laden foam cells, which might be a sign of atherosclerosis pathogenesis. Secondary TW, PP, PE and PS MNP particles impaired cell membrane, decreased cell viability, increased ROS generation, and induced inflammatory and hemolysis responses in human peripheral blood mononuclear cells (PBMCs), murine Raw 264.7 cells, and human mast cell line HMC-1 [48-50, 55]. In contrast to the above findings, PS MNPs were avidly taken up by THP-1-derived macrophages and ascites exudate-derived J774A macrophages but did not cause any significant cytotoxic effects [136, 163].

Overall, in vitro investigations of MNP toxicity in human cells thus far have not identified severe cytotoxic effects. Despite varying degrees of MNP internalization by various cells and a variety of moderate negative effects occurring in an exposure concentration- and duration-dependent manner, in vitro studies generally show no significant signs of frank cellular toxicity, except at very high MNP concentrations that may not be physiologically relevant for most human exposures.

3.3.3. In vivo MNP toxicity studies

Due to their anatomical and physiological similarities relative to human beings, laboratory rodents are relevant albeit imperfect animal models for studying the in vivo biological effects of MNPs. In recent years, a great number of studies have evaluated the adverse effects of MNPs in rodent models via ingestion, inhalation, and other less significant routes. These studies are broadly summarized in Table 3 and are described in more detail below.

3.3.3.1. Ingestion exposure studies

Ingestion is the most likely route of entry for MNPs in humans and animals. Ingested particles may pass through the intestinal barrier to reach the circulatory or lymphatic systems, and from there translocate to multiple organs to cause secondary exposures. The gut, liver, and kidney, representing the major organs involved in uptake, metabolism/detoxification, and excretion, of most toxicants, respectively, were found to accumulate a significantly greater number of particles than other organs after oral administration in mice [37, 92, 164]. Both microsized and nanosized MNPs were also found in the heart, spleen, lung, brain, uterus, testis, ovary, long bones, and blood after exposures ranging from 24 h (acute exposure) to 90 d (chronic exposure) [37, 38, 62, 64, 165-167]. Importantly, ingested MNP particles were shown to be capable of breaking through highly protective biological barriers, including the blood-testis, blood-placenta, and blood-brain barriers, increasing the potential for harm to reproductive, fetal, and neural tissues [62, 64, 168]. In contrast, Stock et al. found PS MNP particles only in gut tissues and in no other organs after ingestion in mice [136]. Such disparities could result from differences in MNP properties, preparation of suspension and resulting colloidal properties, aging, as well as differences in tissue penetration of fluorophore-labelled MNPs and intrinsic stability of fluorophore-particle linkages. To solve these problems, Positron emission tomography, a non-invasive, sensitive and effective method, has been developed to achieve real-time quantitative monitoring of the biodistribution of plastic particles in vivo [169, 170].

Gastrointestinal toxicity:

Because the gastrointestinal tracts are the first to come into direct contact with ingested materials, it is not surprising that gut tissues invariably contain greater numbers of MNP particles than tissues from any other organs in the body following acute ingestion exposures. Following subacute and chronic MNP ingestion exposures, a wide range of toxic effects in the gut have been observed, including histopathological lesions, local inflammation, decreased mucus secretion, compromised intestinal barrier function, tight junction damage, altered transporter functions, and microbiota dysbiosis [94, 136, 155, 165, 171]. Kim et al. suggested that no observed adverse effect level (NOAEL) of weathered PP MNPs with the average size of 85.2 μm as 25 mg/kg bw/day in rats in a 4-week repeated dose oral toxicity study [57]. The gut microbiota also plays an important role in protecting the intestine and the entire organism against infections by exogenous pathogens [172]. Ingested PS and PE MNPs altered the composition and diversity of gut microbiota, leading to gut microbiota dysbiosis [86, 105, 165, 171]. Moreover, since the gut microbial community maintains co-existent and co-evolved relationships with the immune system across the lifespan, gut microbiota dysbiosis may contribute to immune dysregulation [173]. The imbalance of gut microflora was found to be associated with immune disturbance in mice [86], and enrichment of opportunistic pathogens in the gut resulted in the deterioration of the intestinal barrier [94]. Gut microbiota imbalances are also significantly associated with altered fecal bile acid profiles [105]. MNPs absorbed by the gut could lead to systemic exposure and widespread toxicological effects, including circulatory toxicity, hepatotoxicity, endocrine toxicity, developmental toxicity, neural toxicity, renal toxicity, and muscular toxicity (myopathy).

Circulatory toxicity:

As orally administered MNPs have been identified in the liver, kidney, testis and bone tissues, these particles likely enter the multiple peripheral organs and tissues through the blood or lymphatic circulation [64, 69, 167]. Liu et al. identified the accumulation of MNP particles in the serum in mice after 7-day aerosol inhalation exposure of carboxyl-modified or amine-modified PS MNPs with the size of 80 nm [35]. The toxicity of MNPs within the circulatory system itself should not be neglected. Sun et al. found that exposure to 5 μm PS MNPs via oral gavage caused alterations in blood biomarkers, immune responses, and hematopoietic disturbances [167]. Oral exposure to microsized PS MNPs for 90 days was found to cause cardiac fibrosis, characterized by increased levels of cardiac biomarkers in serum, as well as myocardial cell injury and death, and increased myocardial collagen and extracellular matrix deposition in rats [174, 175].

Hepatotoxicity:

MNPs that cross the intestinal epithelium can reach the liver through either the portal circulation (via portal capillaries leading to the portal vein) or the systemic circulation (via intestinal lacteals that lead to the thoracic duct, which drains into the left subclavian vein). The liver is thus one of the most common locations for MNP particle accumulation and is exceeded only by the intestine as the primary site for MNP accumulation [37, 176, 177]. Oral exposure to PS MNPs in mice induced hepatic oxidative stress, inflammation, and cell death, resulting in hepatotoxicity [105, 176-178]. Pathological changes observed in the liver following oral exposures to PS MNPs include alterations in liver weight, fatty liver index, and hepatic function, as well as histological lesions, such as lipid droplet accumulation and inflammation [37, 177, 179, 180]. As the liver is the primary metabolic organ in mammalian organisms, it is perhaps not surprising that metabolic disorders have been identified following oral MNP exposures in experimental animals [37, 165]. 28-day oral exposure of mice to 5 μm PS MNPs at the dose of 2.5 mg/mL caused significant metabolic disturbances that could potentially lead to metabolic reprogramming, a hallmark of carcinogenesis [167]. Oral exposure to ~80 nm PS MNPs has also been found to cause hyperglycemia and abnormal liver lipid metabolism in mice [181]. Moreover, two studies in a DSS-induced colitis mouse model revealed that oral PS MNP exposures aggravated intestinal toxicity and caused secondary liver injuries, including hepatic inflammation and disordered lipid metabolism [182, 183].

Endocrine toxicity:

Reproductive toxicity is one of the most common indicators and results of endocrine toxicity and endocrine disruption. Following 24 h oral exposure to PS MNPs in mice, 4 and 10 μm, but not 0.5 μm PS MNP particles were found in the testis [64]. It has been suggested that damage to the blood-testis barrier induced by MNP exposure might be the cause of male reproductive toxicity [36, 64]. Park et al. suggested the NOAEL of PE MNPs with sizes of 16.9 ± 1.9 μm in drinking water was lower than 60 mg/kg bw/day for oral toxicity and is lower than 15 mg/kg bw/day for reproductive and developmental toxicity in a 90-day repeated dose oral exposure study in mice [184]. Long-term exposure to PS MNPs in mice and rats led to oxidative stress and inflammation in the testis, as well as other structural and functional abnormalities, including testicular atrophy, increased immature sperm cells, decreased sperm quantity and quality, reduced serum testosterone levels, and disturbed fertility [36, 40, 64, 185-187]. Recent studies have also found that 5-week oral exposure to 25 and 50 nm PS MNPs in rats caused the disruption of the hypothalamic-pituitary-testis (HPT) axis, resulting in hormone imbalance and testicular dysfunction [188, 189]. Female reproductive toxicity has also been identified in several studies. Following 5-week oral exposures in female mice, ~0.79 μm PS MNPs were found to accumulate within the ovaries, accompanied by multiple pathological changes, including increased ROS generation, inflammation, mitochondrial dysfunction, structural alterations, and disturbed development and maturation of germ cells in mice [190]. Subchronic exposure to PS MNPs in female rats caused significant reproductive toxicity, including granulosa cell pyroptosis and apoptosis, ovarian fibrosis, altered sex hormone levels, and decreased ovarian reserve capacity [44, 63, 191]. Wei et al. demonstrated that orally administered 5.0–5.9 μm PS MNPs accumulate to a greater degree in the ovaries of female mice than in the testis of male mice, and have a greater impact on fertility in female than in male mice [186]. Finally, in the only study to date on the effects of oral MNP exposure on non-reproductive hormone systems, Amereh et al. found that 5-week oral exposure to 25 and 50 nm PS MNPs in male rats significantly reduced serum triiodothyronine and circulating thyroid hormone levels [188].

Developmental toxicity:

Evaluation of developmental toxicity is considered to be of great importance in determining and assessing the overall toxic characteristic of a chemical. A growing number of studies have revealed that parental exposure to MNPs leads to toxic effects in both parents and offspring [184, 192]. Oral exposure to PE MNPs for 90 days in parental male and female mice, and for 20 days in dams during the lactation period caused tissue damage and immune disorders in dams and altered survival rate, sex ratio, and body weight in offspring [184]. Maternal exposure to PS MNPs in mice during pregnancy and lactation induced hepatic and testicular toxicity in male offspring [193]. Luo et al. found that maternal exposure to PS MNPs caused fatty acid metabolic disorders in F1 offspring [192]. The same authors found that maternal exposure to PS MNPs during gestation and lactation also caused gut dysbiosis and barrier dysfunction in dams, as well as metabolic disturbance in their F1 and F2 generation offspring [194].

Neurotoxicity:

The brain is arguably the most vulnerable organ in any animal organism, and it is generally considered to be protected from exogenous compounds that reach the circulation by the BBB. However, Kwon et al. found that after 7 d oral exposure, microsized PS MNPs translocated through the BBB and accumulated in brain microglial cells, where they induced microglial activation, immune responses, and apoptosis. [62]. 4-week oral exposure to 5 or 20 μm PS MNPs also significantly increased the activity of acetylcholinesterase (AChE) in the liver, considered a potential biomarker of neurotoxicity, which could lead to a decrease in cholinergic neurotransmission efficiency [37]. In contrast, co-exposure to organophosphorus flame retardants (OPFRs) and 0.5–1.0 μm PS MNPs significantly reduced AChE activity and increased levels of other neurotransmitters involved in cholinergic neurotransmission in mice compared to exposure to OPFRs alone [101]. Finally, no significant neurobehavioral alterations were observed in rats fed with 25 or 50 nm pristine PS MNPs for five weeks [195].

Reports on other organ-specific MNP toxicity to date are limited. In a renal toxicity study, PS MNPs were found to accumulate in the kidneys of mice, causing histopathological lesions and alterations in multiple renal function biomarkers after 8-week oral gavage treatments [155]. In a study of MNP effects on skeletal muscle growth and repair in mice, oral exposure to microsized PS MNPs disrupted the regeneration of skeletal muscle by affecting the balance between myogenic and adipogenic differentiation [196]. Oral exposure to 2 μm PS MNPs at 0.4 mg/d 2d/week for 8 weeks was also found to decrease muscle mass and impair lower grip strength in mice [155].

3.3.3.2. Inhalation exposure studies

Inhalation of airborne particles is another important route of MNP exposure. However, only a limited number of studies have investigated the tissue distribution and toxic effects of inhaled MNPs in animals. Single inhalation to PA MNP aerosol increased blood pressure and affected uterine vascular dilation without pulmonary inflammation in rats[197]. Intratracheal instillation of 5 μm PS MNPs for 3 weeks (3 times per week) resulted in accumulation of the MNPs in lung tissue where they caused concentration-dependent pulmonary fibrosis, as indicated by histopathological lesions, increased expressions of fibrotic markers, and increased oxidative stress [198]. Subacute (14 d) inhalation exposure to 100 nm PS MNPs in rats led to concentration-dependent inflammation in the lung accompanied by minor alterations of physiological and biochemical biomarkers [199]. Liu et al. found that 80 nm PS MNPs administered by aerosol inhalation were able to penetrate the BBB to reach the brain in mice [35]. Fournier et al. found that 20 nm PS particles administered by a single intratracheal instillation to pregnant rats were translocated across the pulmonary epithelial barrier to reach the systemic circulation and accumulated in multiple maternal organs, including heart and spleen, as well as in the placenta and multiple fetal organs, indicating maternal-fetal transfer of the inhaled PS MNPs [34]. In contrast, Han et al. reported that 10-45 μm PE MNPs administered by repeated intratracheal instillations to pregnant and lactating mice (from gestational day 9 through postnatal day 7) were found only in maternal lung, intestine and liver, and caused no significant toxicity in either dams or offspring [200]. Thus, it is worth noting that the accumulation of inhaled particles may cause potential risks to vulnerable neonates depending on particle types and sizes.

3.3.3.3. Studies of other exposure routes

Other exposure routes, such as intraperitoneal (IP) injection and intravenous (IV) injection have been employed to investigate the potential toxic effects of MNPs on specific systems. For example, IP injection of 10 μm PS MNPs in mated female mice during the peri-implantation period resulted in reproductive toxicity, including increased embryo resorption rate, decreased uterine blood supply to the placenta, and immune dysfunction [201]. 3 d IP exposures of mice to 23 nm PS MNPs at an environmentally relevant concentration caused neurotoxicity, including cognitive impairment, oxidative stress, and decreased AChE activity in the brain, as well as genotoxicity, evidenced by DNA damage [106]. Although IV injection is considered an artificial exposure and IV exposure of MNPs to humans is very rare, IV administration is the most straightforward exposure route to investigate vascular toxicity. Zhao et al. found that 140 nm gold core-PS shell MNPs accumulated in the livers and spleens of mice after a single IV injection, and that the PS MNPs were retained in those organs for up to 28 days [202]. Vlacil et al. identified interactions of 1 μm PS MNPs with leukocytes and endothelial cells, leading to particle clearance, endothelial activation, monocyte adhesion, and vascular inflammation in mice exposed to the PS MNPs by IV injection [180]. Finally, IV administration of 42 nm PS MNPs to high-fat diet-induced mice caused disturbances of lipid metabolism, hepatic inflammation, and oxidative stress, contributing to hepatic toxicity and the development of hepatic fibrosis [203].

3.4. Systemic, cellular and molecular mechanisms of MNP-induced toxicity

Recent findings have been gradually uncovering the mechanisms of potential toxicity and detrimental effects of MNPs in human cell lines and mouse models. The current literature on the systemic, cellular, and molecular mechanisms underlying MNP toxicity in in vitro cell lines and in vivo rodent models are summarized below and in Figure 3.

Figure 3. A schematic diagram of potential mechanisms underlying MNP toxicity proposed in in vitro and in vivo studies.

A. Systemic levels. MNPs can be exposed through three routes, including ingestion, inhalation and possible dermal contact. Ingested or inhaled MNPs with small sizes could reach the systemic or portal circulation, and further distribute and accumulate in different organs, leading to organ dysfunction. Moreover, MNP particles could translocate through physiological barriers, including blood-placenta barrier, blood-brain barrier and blood-testis barrier, posing a potential risk to neural, reproductive and developmental systems. B. Cellular and molecular levels. The internalized MNP particles can cause lysosomal damage or mitochondrial dysfunction, which induces reactive oxygen species (ROS) production. Oxidative stress caused by overproduced ROS leads to a cascade of cellular effects, including endoplasmic reticulum (ER) stress, cytoskeletal dysfunction, protein oxidation, lipid peroxidation, and DNA damage. Moreover, oxidative stress leads to activation of molecular signaling pathways, which could further activate inflammation and transcription factors. These interconnected intracellular events collectively contribute to cell injury and death.

3.4.1. Systemic mechanisms

As described in more details above, studies in animal models have shown that some MNPs can be taken up by and translocate across the gastrointestinal or respiratory epithelium following ingestion and inhalation exposures, respectively, to enter the portal or systemic circulation. Following uptake, MNPs can be widely distributed to organs via the systemic circulation and in some cases can bypass physiological barriers to enter the tissues. Several mechanisms of systemic MNP toxicity have been identified in vivo, including gut dysfunction, pulmonary and systemic inflammation, hepatic metabolism disorders, tissue fibrosis, reproductive toxicity, and neurotoxicity. While such in vivo studies provide critical information for risk assessment, in vitro studies help us understand the cellular and molecular mechanisms underlying MNP toxicity and may provide insights into possible approaches for protecting against or minimizing the toxic effects induced by MNP particles.

3.4.2. Cellular and molecular mechanisms

Particle internalization:

A number of studies have identified possible pathways of cellular uptake of MNPs. Bonanomi et al. showed that both ATP-dependent and ATP-independent processes were involved in the uptake of both micro-sized and nano-sized PS MNPs in human colon CCD-18Co cells after acute or chronic exposure [73]. Ding et al. found that the uptake of 60 nm PS MNPs by human GES-1 cells was significantly decreased by inhibition of caveolae-mediated and clathrin-mediated endocytosis as well as micropinocytosis [38]. DeLoid et al. observed actin shells enveloping 25 nm PS MNPs during internalization by Caco-2 enterocytes in a triculture small intestinal epithelial model, suggesting the involvement of actin-dependent endocytic pathways (macropinocytosis or phagocytosis) in MNP uptake. These investigators further found that uptake was decreased following inhibition of dynamin with Dyngo, suggesting possible contributions of dynamin-dependent endocytic pathways (clathrin-mediated and fast endophilin-mediated endocytosis) [46]. Xu et al. found that dynamin inhibition with dynasore and specific inhibition of clathrin-mediated endocytosis with chlorpromazine, as well as inhibition of macropinocytosis with amiloride, significantly decreased uptake of 100 nm PS MNPs by Caco-2 cells, suggesting involvement of both clathrin-mediated endocytosis and macropinocytosis [65]. Other studies have shown that MNPs can be taken up and released by cells via ATP-independent mechanisms. Fiorentino et al. found that 44 nm PS MNPs were internalized through ATP-independent processes in bovine oviductal epithelial cells and human colon fibroblasts [204]. Likewise, Liu et al. showed that 50 and 500 nm PS MNP are taken up by rat basophilic leukemia (RBL-2H3) cells via both passive membrane diffusion and active endocytosis mechanisms, whereas 5 μm PS MNPs were not taken up, likely due to their large size and weak Brownian motion [205]. The mechanisms involved in cellular MNP uptake depend on various factors, including MNP particle size, charge and hydrophobicity, cell types, and culture medium. Particle size is a crucial determinant of MNP internalization, with smaller nano-scale MNPs more readily taken up than larger micro-scale MNPs [46]. Moreover, specific mechanisms involved in MNP uptake are particle size-dependent. Liu et al. found that uptake of 50 nm PS MNPs occurred via clathrin- and caveolae-mediated pathways, whereas uptake of 500 nm PS MNPs occurred by micropinocytosis, although as noted above, MNPs of both sizes were also taken up by passive diffusion [205]. Similarly, 0.2 and 2 μm but not 10 μm PS MNPs were found to be taken up by microglial cells via phagocytosis, suggesting that phagocytosis of MNPs by immune cells is size-limited [62]. Surface charge also affects the cellular uptake of particles as evidenced by the findings that positively charged MNPs were more readily internalized by nonphagocytic cells, whereas phagocytic cells preferentially took up anionic MNPs [206].

Lysosomal damage:

Internalization of particles via endocytic processes initially generates primary endocytic vesicles, which subsequently deliver their cargo and membranes to early endosomes in the outer cytoplasm. Early endosomes then move along microtubules toward the perinuclear space as they acquire additional membrane components and lysosomal hydrolases, acidify, and undergo homotypic fusions to become mature lobulated late endosomes. Late endosomes subsequently fuse with lysosomes to form endolysosomes, where degradation of the cargo is carried out by lysosomal hydrolases, including proteases, nucleases, glycosidases, phosphatases, phospholipases, and sulfatases. Following cargo degradation, endolysosomes are converted to dense lysosomes that serve as storage depots for lysosomal enzymes and membrane components [207]. However, as none of the known lysosomal hydrolases are capable of hydrolyzing plastic polymers, MNPs internalized by endocytosis and arriving in endolysosomes would presumably not be significantly degraded, and the MNP cargo of the endolysosome would then persist and accumulate in the endolysosomes and dense lysosomes. Excessive accumulation of MNPs within the lysosomal compartment could suppress lysosomal activity, induce lysosomal instability and rupture, and release MNPs and lysosomal hydrolases into the cytoplasm [42, 208, 209]. MNP particles so released from damaged lysosomes or that entered the cytoplasm through passive diffusion could also target and enter mitochondria and other organelles, facilitated by oxidative damage to those organelles. However, protein corona covering MNPs under a biological microenvironment protected cells from intracellular membrane damage, lysosomal dysfunction, cell apoptosis and autophagy until corona digestion [210, 211].

Mitochondrial dysfunction:

Mitochondrial dysfunction and damage caused by MNP exposure not only serve as the primary source of ROS generation but also induce cell apoptosis. Metabolic analysis of livers from mice following 28 d oral gavage exposures to 5 μm or 20 μm PS MNPs revealed dose-dependent disturbances of energy metabolism, evidenced by significantly decreased ATP levels and increased LDH activity, and suggesting mitochondrial dysfunction [37]. In Caco-2 cells, PS MNPs induced mitochondrial swelling and depolarization, which could ultimately promote mitochondrial damage and rupture with subsequent activation of cell death pathways [67, 135]. Exposure of GES-1 cells to chlorinated PS MNPs caused mitochondrial depolarization via particle-induced oxidative stress, subsequently leading to cell apoptosis [97]. Moreover, oral exposure to PS MNPs in rats induced mitochondrial damage and dysregulation of mitochondrial potential in cardiomyocytes, which further promoted ROS generation and oxidative stress, leading to NOD-like receptor protein 3 (NLRP3) inflammasome and caspase-1 activation, ultimately resulting in apoptotic and pyroptotic cell death and cardiac dysfunction in mice [175]. This sequence of events, from mitochondrial damage to cell death, was also found to contribute to liver injury in mice following PS MNP exposures [177].

Oxidative Stress:

Oxidative stress serves a pivotal role in cellular responses to MNPs, leading to further cytotoxic and inflammatory effects. Oxidative stress is characterized by an imbalance between ROS generation and the level of antioxidant defenses. Overproduction of ROS caused by particle-induced mitochondrial or lysosomal damage can exceed the antioxidant defense capacity of cells, leading to oxidative damage at the cellular or molecular levels. A putative Adverse Outcome Pathways (AOP) analysis based on reports of MNP toxicity mechanisms identified ROS formation as the molecular initiating event leading to adverse outcomes due to MNP exposures [212]. Multiple studies have identified indicators of oxidative stress induced by MNPs, including increased production of ROS and malondialdehyde [146, 155, 177, 191], and decreased levels of antioxidant enzymes such as glutathione peroxidase, catalase and superoxide dismutase [40, 44]. Moreover, it was recently found that cyanidin-3-O-glucoside, a bioactive anthocyanin compound with anti-oxidative, anti-inflammatory, and gut microbiota modulating effects, alleviated PS MNP-induced oxidative stress and inflammation, promoted particle excretion, and restored gut microbiota homeostasis in mice [213]. Likewise, NaHS, an H₂S donor with antioxidant and free radical scavenging properties, has been shown to protect against PS MNP-induced hepatotoxicity by ameliorating oxidative stress [176, 214]. Particle-induced ROS can nonspecifically interact with biomolecules such as proteins, lipids, and nucleic acids to cause further cell injury. For example, lipid peroxidation, caused by the reaction of lipids with cellular ROS, leads to membrane disruption and lipid dysmetabolism [91, 162, 181]. Oxidative damage to DNA (DNA base oxidation) can also have serious adverse consequences. If not efficiently repaired with high fidelity, such DNA damage can lead to transcription/replication errors, mutations, cellular dysfunction, or eventually cell death [215]. These findings suggest that genetic variation in susceptibility to MNP exposures may also exist. Recent studies have identified genotoxic stress and DNA damage caused by MNP exposure as possible genotoxic mechanisms underlying MNP toxicity [103, 216]. In addition, intracellular ROS can interact with mitochondrial DNA, subsequently leading to mitochondrial dysfunction [217]. Oxidative stress could also explain MNP-induced perturbations of intracellular calcium homeostasis. Li et al. found that PS MNP-induced ROS overproduction promoted the activity of store-operated calcium entry channels, leading to intracellular calcium overload [218]. Such accumulation of intracellular calcium levels can exert toxic effects on cells via multiple pathways.

Inflammation:

Inflammation is considered one of the primary oxidative stress-induced cellular events. Oxidative stress triggered by MNP exposure activated inflammation signaling pathways, including nuclear factor kappa B (NF-κB) and mitogen-activated protein kinase (MAPK) pathways [185, 219], which are associated with the enhanced secretion of proinflammatory cytokines, including tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), and interleukin-8 (IL-8) [48, 51, 85].

Cytoskeleton dysfunction:

The cytoskeleton is not only susceptible to direct attack by intracellular xenobiotic particles but is also a primary target of ROS [220]. PS MNPs can cause cytoskeletal dysfunction, characterized by decreased expressions of cytoskeletal proteins, including α-tubulin and disheveled-associated activator of morphogenesis (DAAM-1) in rat ovaries. Moreover, DAAM-1 protein is an important component of the non-canonical Wnt signaling pathway, the inactivation of which may contribute to female infertility [191]. PS MNPs were also found to cause particle-induced oxidative stress-mediate upregulation of the canonical Wnt/β catenin pathway, leading to ovary fibrosis [63].

Activation of signaling pathways:

ROS also serve as second messengers for activating a complex network of signaling pathways involved in energy metabolism, inflammation, cell survival, and cell death. ROS production triggered by PS MNP exposure activated the AMP-activated protein kinase (AMPK)-PGC-1α signaling pathway, resulting in glycolysis imbalance [218]. Oxidative stress caused by exposure to PS MNPs was also shown to activate MAPK signaling pathways, contributing to testicular toxicity in rats [36] and mice [40]. Excessive ROS production leading to activation of the MAPK/ hypoxia-inducible factor (HIF1α) pathway was identified as the mechanism for PS MNP-induced apoptosis and necroptosis in swine testis cells [221]. PS MNP-induced oxidative stress was also shown to provide a priming signal for the NLRP3/Caspase-1 pathway mediating cell apoptosis and pyroptosis in rat ovarian granulosa cells [44].

Cell injury and death:

Oxidative stress and related intracellular pathways contribute to cell injury and death. In addition, MNP-induced oxidative stress was found to promote the phosphorylation of ZO-1 protein, leading to the disruption of intercellular tight junction [144]. MNPs have also been shown to induce different types of cell death, including autophagy [155], apoptosis [222], necrosis, necroptosis [221], pyroptosis and ferroptosis [177] in a variety of in vitro cell systems and in vivo animal models depending on the concentration and duration of exposure. Dead or dying cells not promptly recognized and removed by macrophages can lead to further inflammatory responses and tissue injuries [223]. Induction of cell death mechanism by MNP particles might serve as the cornerstone of multiple associated pathologies. Oxidative stress and cell death are considered the dominant mechanisms for MNP-induced tissue injuries and organ dysfunction [44, 63, 174, 175]. Other cellular mechanisms underlying MNP toxicity such as activated endoplasmic reticulum (ER) stress [155] and inhibited activity of ABC transporters [41, 67] have also been reported. Further research is needed to elucidate the details and contributions of specific mechanisms to MNP-induced cell injury and death.

In summary, the multiple intracellular mechanisms underlying MNP toxicity are interdependent, and oxidative stress is the torchbearer for many of them. ROS have important cellular roles, including inducing multiple intracellular events and acting as second messengers for the activation of specific signaling pathways and gene expressions. However, the various studies investigating potential mechanisms underlying MNP toxicity have each employed MNPs of only one polymer (most often PS), and cannot be extrapolated to MNPs of other polymers. Further studies are needed across the full range of common MNP polymers to close the remaining knowledge gaps regarding the toxicity mechanisms of other types of MNPs.

4. Discussion

Plastics are increasingly and widely used in all aspects of our daily life. However, our understanding of the potential hazards of MNPs produced from the degradation and fragmentation of plastic waste lags behind their increasing prevalence in the environment and food web. The small size and ubiquitous nature of MNPs in the environment pose multiple challenges for designing and performing studies to examine their adverse health effects in humans. It has become apparent that we need to assess potential health implications across the life cycle of plastic materials to include MNPs. Based on our current understanding and knowledge of MNP toxicity and hazards, we have identified several knowledge gaps and future research needs.

4.1. Knowledge gaps

4.1.1. MNP types and physico-chemical properties

The vast majority of MNP toxicity studies to date have been limited to commercially available primary PS MNPs, with a few studies employing commercially available primary MNPs of a limited number of other polymers such as PE and PP. These manufactured primary MNP particles have spherical shapes, uniform sizes, and smooth pristine surfaces. In contrast, the MNPs found in the environment and food web, to which humans are far more likely to be exposed, are secondary MNPs of many different polymers. Secondary MNPs often contain chemical additives and are generated by the degradation of plastic waste in the environment or by municipal incineration of plastic waste. Such environmentally relevant MNPs tend to have irregular shapes, broad size distributions, and rough surfaces chemically modified by photo-oxidation and other environmental interactions [224]. Because MNPs with different physicochemical properties can have significantly different toxicity profiles [225], the results of toxicity studies employing pristine primary MNPs would differ substantially from the results of studies employing environmentally relevant MNPs. Moreover, human exposure to MNPs is complicated by interactions of MNPs with physiological fluids and processes in the human body, which cause further MNP transformations before they interact with cells [46, 111]. We are thus far from a comprehensive understanding of the roles played by MNP polymer types and other physicochemical properties in their interactions with biological systems and toxicity, which thus represents a major knowledge gap.

In addition, many of the plastics from which environmental MNPs are derived contain chemical additives, which can leach from MNPs in the environment or within the body or cells during exposure. There is also growing evidence that MNPs can adsorb and concentrate other environmental pollutants (EPs), including heavy metals, pesticides, POPs, and chemicals leached from MNPs themselves [226, 227]. This could represent an additional potential health hazard through either the desorption of EPs in the human body to increase their bioaccessibility or bioavailability, or by impairing physiological barrier function to further increase EP bioavailability [228]. In contrast, deposits of MNPs within the body may present relatively stable adsorption sinks for some chemicals, which could have additional adverse long-term effects. Co-exposures to MNPs and adsorbed EPs can have synergistic, additive, or even antagonist effects [229, 230]. The effects of EP adsorption/desorption and leaching of additives on MNP toxicity thus represent a critical knowledge gap for risk assessors.

4.1.2. MNP exposure concentrations and durations

Exposure concentrations and durations are critical determinants of toxicity for any chemical or particle, including MNPs. Some studies reviewed here reported concentration- and duration-dependent toxicity of MNPs. However, most MNP toxicity studies to date have employed high-dose and short-term exposures, which do not accurately reflect the low-dose and long-term exposures to MNPs that humans would experience in the real world.

4.1.3. Toxicity studies

Although extensive in vitro testing, animal studies, and mathematical models have shown that MNP exposures may pose significant health risks to humans, there is currently no direct measurable evidence of human health risk. In vitro toxicity studies are used to help identify and characterize the intrinsic toxicity of substances and provide data for establishing in vivo exposure to those substances. However, most in vitro MNP toxicity studies to date are limited to traditional monocultures of immortalized cell lines, which do not approach the structural or functional complexity and thus physiological relevance of in vivo tissues, much less whole organisms. Some studies have employed 3D in vitro cell culture models incorporating multiple cell types to provide more complete and physiologically accurate representations of human tissues or biological barriers. For example, the cell model most commonly used to investigate MNP uptake and toxicity in the intestine is the matured Caco-2 monoculture, representing a monolayer of naked absorptive intestinal enterocytes. However, in the human intestinal enterocytes are not naked—and thus exposed directly to luminal contents, but are sheathed in a layer of mucus produced by goblet cells, which make up 4-16% of intestinal epithelial cells [231]. The mucus layer is an integral component of the intestinal barrier and could either restrict access of ingested MNPs to enterocytes or alter the coronae of MNPs as they traverse it, thereby modulating subsequent interactions with the cells beneath the mucus. In order to take the mucus layer into account, some recent studies have employed co-cultures of Caco-2 and mucus-secreting goblet cell-like HT29-MTX cells, as discussed in the section on cellular models above. To provide more meaningful and predictive data, future in vitro studies of MNPs should employ cellular models that provide greater structural and physiological consistency with the corresponding in vivo systems. This could reduce the occurrence of contradictory results between in vitro and in vivo studies like those often seen for MNP absorption. In addition, studies focusing on toxicity at the initial absorption site, secondary systemic signaling, biodistribution and bioaccumulation, and local perturbations in peripheral target tissues are limited. There is also a lack of subchronic and chronic exposure studies in animal models that have addressed the total absorption, distribution, metabolism, and excretion (ADME) of MNPs. It is also worth noting that the cumulative intake and flux of MNPs in humans arise from multiple exposure routes and include a variety of MNP species. No studies to date have addressed this mixed MNP situation.

Large research gaps also exist in our understanding of the effects of dermal and ocular MNP exposures. Though these exposure routes may not significantly contribute to systemic MNP exposures, localized dermal and ocular effects of MNPs could have serious health consequences. Moreover, data from in vitro and in vivo studies of these exposure routes to date have produced contradictory results. For example, studies of MNP exposures in 3D human skin and cornea culture models revealed no signs of skin or eye inflammation or irritation [57], whereas exposure of the ocular surface in mice to MNPs resulted in significant xeropthalmia-like ocular surface injuries [149].

In addition, studies of potential genotoxicity, mutagenesis, and carcinogenicity of MNPs are scarce and incomplete. Genotoxicity studies to date are limited to screening for genotoxicity via the traditional Comet assay in monoculture cell line models [103, 161, 232, 233]. MNP exposures have been found to cause DNA damage, oxidative stress, chronic inflammation, and angiogenesis, all of which are associated with carcinogenicity. Yet to date, no studies of MNP mutagenicity or carcinogenicity have been reported. Such studies are time-consuming and resource-intensive, and in the case of carcinogenicity require lifetime chronic exposure studies in animals. However, given that we know that chronic human MNP exposures are ongoing and likely to increase, the expenditure of time and resources required to perform these studies is certainly warranted.

4.1.4. Potential mechanisms underlying MNP toxicity

Mechanisms underlying the absorption, distribution, accumulation, and organ-specific toxicity of MNPs based on animal studies to date are still unclear. At the cellular level, in vitro studies have shown that MNPs can be internalized via both energy-dependent and -independent pathways. However, these studies have most often employed monoculture cellular models lacking physiological relevance and pristine primary PS MNPs lacking environmental relevance. The lack of data on the relationships between uptake pathways and MNP polymer types and physicochemical properties thus represents a large and important knowledge gap.

Oxidative stress has been shown to play a key role in most mechanisms underlying MNP toxicity [74, 220, 221, 223, 234, 235]. However, there is as yet little known about the specific relationships between oxidative stress and subsequent intracellular events responsible for MNP toxicity. Studies to date have been limited to specific molecular signaling pathways and have not clarified the interactions between the various signaling pathways involved in MNP toxicity. The remaining gaps in our understanding of the molecular events initiated by MNP exposure could impede efforts to identify early biomarkers for MNP toxicity in humans.

4.2. Research needs

To better understand the toxic impacts of MNPs on humans, we would suggest that future research focus on the following areas to address the knowledge gaps identified in this review: 1)Assessing the biokinetics and toxicity of environmentally relevant MNPs, with surface chemistries and other physico-chemical properties consistent with photo-oxidative degradation in the environment or incineration of plastics, at environmentally relevant concentrations, across the full range of highly produced polymers, using physiologically relevant cellular and animal models; 2) Determining the adsorption and desorption of EPs from MNPs and assessing the impacts of MNPs on the bioaccessibility, bioavailability, toxicity, and biokinetics of these chemicals pollutants—using environmentally relevant MNPs across the range of polymers and physiologically accurate biological models; 3) Systematically investigating the cellular and molecular mechanisms underlying toxicity of environmentally relevant MNPs (using relevant cellular and animal models) to understand the adsorption, accumulation, translocation, and metabolism of MNPs in the human body and to identify potential pathways and biomarkers for MNP toxicity.

It is also important to note that the presence of particles, such as MNPs, in experimental systems can often interfere with measurements of absorbance, fluorescence, or luminescence that are the basis of many toxicological assays, which could cause significant errors and misinterpretation of results. Another potential source of error in particle studies is leeching or desorption of chromophores or fluorophores from labeled particles. To avoid such potential errors, future studies of MNPs must include appropriate controls to rule out or quantify potential interference, and to account for potential leeching or desorption of chromophores or fluorophores from labeled MNPs. More studies on MNP toxicity along the lines suggested above are clearly needed to provide environmentally and physiologically relevant data that can provide a rational basis for developing appropriate science-based risk assessment models to evaluate the range and extent of human health hazards posed by increasing contamination of our environment, food, and beverages by MNPs. A comprehensive toxicological assessment of MNPs should be undertaken post-haste to provide the data and evidence needed by regulators and lawmakers to advance public health policies to regulate plastics manufacturing and use as needed to limit and reduce human exposures and their potential adverse health consequences.

5. Conclusions

The potential toxicity of MNPs is gaining considerable interest as a result of our growing awareness of their ubiquitous and increasing presence in the environment, throughout the food web, and even within our own tissues. There is ample evidence that we are all already harboring MNPs in our blood and other tissues. Numerous studies reviewed above have been conducted to better understand the impact of MNPs on environmental and human health. MNP toxicity is greatly dependent on the MNP physico-chemical characteristics, which can be altered by various environmental factors and physiological processes. In vitro and in vivo studies reviewed above suggest that MNP toxicity and bioactivity are largely determined by physico-chemical characteristics of these particles. However, most of MNP toxicity studies to date used commercially manufactured primary MNPs instead of environmentally relevant secondary MNPs. We also summarized results from MNP toxicity studies that have begun to identify the cellular and molecular mechanisms underlying MNP toxicity, which might provide insights into possible interventions to prevent or mitigate MNP toxicity. Further research is needed to bridge critical knowledge gaps in our current understanding of MNP toxicity and potential health risks of these particles, and provide accurate and complete information for researchers, regulators, lawmakers, and the public. This information will help address the interrelated ecological and public health implications of massive and exponentially increasing production and use of plastics and the consequent substantial and increasing contamination of our environment and food by MNPs.

Highlights.

• Micro- and nanoplastics (MNPs) generated from degradation and fragmentation of plastic waste pose great threats to environment and human health due to their small sizes and ubiquitous distribution.

• MNPs used in most toxicity studies to date were commercially available primary MNPs, rather than environmentally relevant secondary MNPs.

• In vitro and in vivo evidence reviewed here suggests that MNP toxicity and bioactivity are largely determined by MNP particle physico-chemical characteristics.

• Oxidative stress appears to serve as the key event in MNP toxicity, with the excess ROS generated then interacting with other cellular processes to produce cytotoxicity.

• Large gaps in our knowledge of MNP toxicity and the potential health impacts of MNP exposures remain, including the accurate understanding of human exposures and the toxicity and health impacts of environmentally relevant secondary MNPs at relevant exposure doses and durations, and much further study is needed to bridge those gaps.