Manganese Dioxide Nanosheets Induce Mitochondrial Toxicity in Fish Gill Epithelial Cells

Abstract

The development and production of engineered 2D nanomaterials are expanding exponentially, increasing the risk of their release into the aquatic environment. A recent study showed 2D MnO₂ nanosheets, under development for energy and biomedical applications, dissolve upon interaction with biological reducing agents, resulting in depletion of intracellular glutathione levels within fish gill cells. However, little is known concerning their toxicity and interactions with subcellular organelles. To address this gap, we examined cellular uptake, cytotoxicity and mitochondrial effects of 2D MnO₂ nanosheets using an in vitro fish gill cell line to represent a target tissue of rainbow trout, a freshwater indicator species. The data demonstrate cellular uptake of MnO₂ nanosheets into lysosomes and potential mechanisms of dissolution within the lysosomal compartment. MnO₂ nanosheets induced severe mitochondrial dysfunction at sub-cytotoxic doses. Quantitative, single cell fluorescent imaging revealed mitochondrial fission and impaired mitochondrial membrane potential following MnO₂ nanosheet exposure. Seahorse analyses for cellular respiration revealed that MnO₂ nanosheets inhibited basal respiration, maximal respiration and the spare respiratory capacity of gill cells, indicating mitochondrial dysfunction and reduced cellular respiratory activity. MnO₂ nanosheet exposure also inhibited ATP production, further supporting the suppression of mitochondrial function and cellular respiration. Together, these observations indicate that 2D MnO₂ nanosheets impair the ability of gill cells to respond to energy demands or prolonged stress. Finally, our data demonstrate significant differences in the toxicity of the 2D MnO₂ nanosheets and their microparticle counterparts. This exemplifies the importance of considering the unique physical characteristics of 2D nanomaterials when conducting safety assessments.

Introduction

In recent years the discovery and application of emerging two-dimensional (2D) materials, or “nanosheets”, has increased exponentially. Various metal oxides, hydroxides, chalcogenides and elemental allotropes may exist in the form of ultrathin 2D “nanosheets” (Tan et al. 2019) and these nanosheets can be produced by exfoliation of bulk layered crystals or bottom-up assembly into atomically thin layers. 2D layered materials consist of multiple covalently-bonded sheets or plates held together in stacks by van der Waals and in some cases electrostatic forces between the individual layers (Ahmadi et al. 2020). These materials often have micron-scale lateral dimensions (in the layer plane) but nano-scale thickness. This geometry is associated with large surface area and surface-to-volume ratio relative to bulk materials, and the phenomenon of 2D electron confinement, atomically thin edges, edge reactivity and flexibility (the ability to adopt wrinkled or folded conformations) result in novel properties that distinguish them from their more common 0D (particulate) counterparts (Ahmadi et al. 2020, Choi et al. 2017). These novel properties make 2D nanosheets interesting for diverse applications relevant to energy technologies, electronics, water treatment, and medical devices and therapies (Ahmadi et al. 2020).

Engineered nanomaterials can enter the environment at any stage during their life cycle (Gottschalk and Nowack 2011, Sun et al. 2017). Currently, production processes and facilities, wastewater treatment plants and accidents during transport have been identified as the main sources of environmental release of engineered nanomaterials (Farre et al. 2009). As 2D nanosheets are already being incorporated into products across a wide variety of industries (Tan et al. 2019), environmental release of this emerging class of materials is inevitable. Therefore, it is crucial to understand how the unique shape, surface area and stability or dissolution behaviour of these materials influence their toxicity.

Chemical transformation and dissolution pathways and kinetics are key determinants of 2D nanomaterial toxicity (Gray et al. 2018, 2020, Tan et al. 2019). The unique structure, shape and high surface area of 2D nanosheets can impact their rate of dissolution in environmental and biological media. A case study of MnO₂ nanosheets for example, showed that this nanomaterial is stable in pure water, but undergoes dissolution in the presence of common biological reducing agents, releasing toxic manganese ions (Mn²⁺) (Gray et al. 2020). The 2D MnO₂ nanosheets had significantly faster dissolution kinetics than their MnO₂ microparticle counterparts (Gray et al. 2020), exemplifying the importance of shape and/or surface area in determining the dissolution behavior of the 2D nanosheets. Since MnO₂ nanosheets undergo reductive dissolution (involving electron donors that convert Mn(IV) to Mn(II)), they are likely to be stable in some aquatic environments, and undergo dissolution after being taken up by a target organism containing physiological reducing agents, both extracellular and intracellular.

Fish serve as a valuable indicator species for both the detection of contaminated freshwater ecosystems and identifying potential toxicity of environmental pollutants (Lopez and Sedeno-Diaz 2015). Fish gills are an initial target site of exposure to aquatic environmental pollutants and accumulate nanomaterials through their association with mucous proteins on the surface of the gill (Smith et al. 2007, Evans et al. 2005). The well-characterized rainbow trout cell line, RTgill-W1, was derived from rainbow trout gill epithelial tissue, providing an alternative to animal toxicity testing and to further mechanistic toxicity studies (Bols et al. 1994). The RTgill-W1 cell line expresses xenobiotic metabolizing enzymes and membrane transporters and has been validated for aquatic toxicity testing (Nimmo et al. 1985, Eyckmans et al. 2011, Lee et al. 2009). This model revealed that MnO₂ nanosheets were taken up by fish gill cells prior to dissolution, resulting in decreased levels of the glutathione, an important intracellular reducing agent (Gray et al. 2020).

The observed depletion of cellular glutathione suggests intracellular reductive dissolution of the MnO₂ nanosheets and subsequent release of manganese ions (Gray et al. 2020). Manganese is a transition metal with multiple oxidation states, of which Mn³⁺ and Mn²⁺ are most common in biological systems (Michalke et al. 2007). It is an essential trace element involved in antioxidant activity, iron metabolism, brain function and bone formation (Michalke et al. 2007, Michalke and Fernsebner 2014). However, an excess of manganese is toxic, resulting in acute neurological defects. Mitochondria are the major target of manganese toxicity where it inactivates key enzymes involved in oxidative phosphorylation, metabolism and calcium and iron homeostasis (Chen 2018, Harischandra et al. 2019, Michalke and Fernsebner 2014, Roth and Garrick, 2003).

Manganese accumulation in fish results in the generation of reactive oxygen species, oxidative damage and impairment of mitochondrial activity (Crawford et al. 2011, Dolci et al. 2013, Doci et al. 2014, Vieira et al. 2012). Additionally, manganese disrupts oxygen utilization in oyster gill mitochondria (Crawford et al. 2011). Since manganese impairs mitochondrial activity and function in a variety of aquatic species, it is important to understand whether MnO₂ nanosheets induce mitochondrial toxicity in fish gill cells. Thus, the main goals of this study are (1) to elucidate the mechanisms of intracellular uptake and dissolution of MnO₂ nanosheets and (2) to determine the effect of MnO₂ nanosheets on gill mitochondrial integrity and function. Additionally, we compare the mitochondrial toxicity of the 2D MnO₂ nanosheets with MnO₂ microparticles in order to determine whether the unique physical characteristics of the nanosheets influence their toxicity.

Materials and Methods

Manganese oxide materials

MnO₂ nanosheets were synthesized using an adaptation of the aqueous-phase bottom-up method described by Liu et al. (2015). Briefly, this method assembles nanosheets by reduction of KMnO₄ precursor and nucleation on sheet-like dodecanol templates derived in situ from hydrolysis of the surfactant sodium dodecyl sulfate (SDS). A large round bottom flask containing 2.2 L of nanopure water was heated to 95°C and SDS and H₂SO₄ aqueous solutions were added to the flask resulting in concentrations of 0.23 M for each compound. After 15 min, KMnO₄ was added to the mixture, producing an overall concentration of 17.9 mg/L. This mixture was maintained at 95°C for 60 min. The reaction was quenched by filtering the MnO₂ product using vacuum filtration through a 0.22 μm sterile vacuum filter. The product was washed six times alternating between 100% ethanol and autoclaved, nanopure water. Stock solutions of MnO₂ microparticles (Sigma Aldrich 310700) were prepared in 100% ethanol and used as a microparticle reference material. Both the nanosheets and microparticles were stored as aqueous suspensions at 4°C.

The MnO₂ nanosheets and microparticles have previously been fully characterized for size, zeta-potential and agglomeration behavior in cell culture media (Gray et al. 2020). A summary of these data is provided in Tables S1a, S1b and Figure S1. Briefly, the MnO₂ nanosheets have lateral dimensions ranging from 100–300 nm and similar X-Y (in-plane) dimensions. The nanosheets are a mixture of monolayer and few-layer sheets, ranging in thickness from 1–3 nm. The MnO₂ microparticles are round in shape and have an average diameter of 300 nm. Both the MnO₂ nanosheets and microparticles displayed negative surface charge across a pH range of 3–10. Both the nanosheets and microparticles showed little agglomeration and similar colloidal stability in cell culture media (Leibovitz’s L15, RPMI and EMEM media) over the course of 72 h (Gray et al. 2020).

In vitro cell culture and manganese exposure

The well-characterized fish gill cell line RTgill-W1 (ATCC CRL-2523), derived from rainbow trout gill epithelium (Oncorhynchus mykiss) (Bols et al. 1994) was utilized for this study. RTgill-W1 cells were cultured in Leibovitz’s L15 medium (ThermoFisher Scientific 31415029; Table S2) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin, maintained at 19⁰C in a cooling incubator and sub-cultured weekly in T75 cell culture flasks. The formulation of the L15 medium is described in Table S2. Cells used in experiments were between passages 3–25. For all experiments, cells were seeded in monolayer in either 96 well or 6 well plates and allowed to rest for 48 h before being exposed to manganese compounds. Stock solutions of MnO₂ nanosheets and MnO₂ microparticles suspended in sterile water at 1.5 mg/mL were bath sonicated for 1 h. MnO₂ materials were then diluted to 224 μg/mL in L15 medium and sonicated for 15 min before final concentrations of 1.6–160 μg/mL were prepared in L15 medium. Manganese (II) chloride tetrahydrate (Sigma Aldrich M3634) was dissolved in sterile water at 500 mM and final concentrations of 1–100 ppm were prepared in L15 medium. Experiments that included MnCl₂ as a soluble Mn control were conducted at equal ionic Mn concentrations between MnCl₂ and MnO₂ materials (assuming 100% dissolution). Comparisons of the particulate and soluble Mn exposures are outlined in Supplemental Table S3. Experiments were completed in Leibovitz’s L15 medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin, which is required for the normal growth and health of the RTgill-W1 cells. This eliminated any additional stress factors potentially introduced through a change in media composition, which allowed us to investigate nanosheet toxicity and sub-cytotoxic effects.

Lysosomal dyes and fluorescent microscopy

Colocalization of MnO₂ nanosheets and microparticles with lysosomes was visualized using fluorescent stains and confocal microscopy. Following exposure to 8.7 ug/mL MnO₂ nanosheets or microparticles, lysosomes and nuclei were fluorescently stained with LysoTracker Red DND 99 (250 nM; Invitrogen L7528) and Hoechst 33342 (1:3000, Invitrogen H3570) respectively for 30 min at room temperature. Confocal images of live cells were obtained with a Zeiss LSM 710 confocal microscope at 63x. Images were presented as maximum intensity projections.

Intracellular cathepsin B localization was utilized as a marker of lysosomal integrity. After exposure to MnCl₂, MnO₂ nanosheets or microparticles (50 ppm Mn ion; 79.1 ug/mL MnO₂), cells were washed with PBS and incubated in Magic Red in vitro Cathepsin B reagent (1:32 final dilution, ImmunoChemistry Technologies) and Hoechst 33342 (1:10,000, Invitrogen H3570) for 30 min at room temperature. Cells were washed with PBS and imaged live on an Olympus IX81 microscope with a 63x objective.

Transmission electron microscopy

Transmission electron microscopy was used to visualize internalized MnO₂ nanosheets and microparticles and visualize mitochondrial morphology following 24 or 48 h exposure to 8.7 ppm MnO₂ nanosheets or microparticles. At harvest, samples were fixed in 2.5% glutaraldehyde, 2% paraformaldehyde and 2 mM calcium chloride in 0.15 M sodium cacodylate buffer for 2 h. Samples were post fixed with 2% osmium tetroxide/potassium ferrocyanide mixture, 1% thiocarbohydrazide, then 2% osmium tetroxide with appropriate rinses between staining steps. Specimens were then stained with 1% uranyl acetate overnight, followed by Walton’s lead aspartate (pH 5.5) at 60°C for 30 min. Finally, samples were dehydrated in graded ethanol solutions, infiltrated with Araldite Embed 812 embedding medium and cured at 60°C for sectioning. Micrographs were obtained on a Philips 410 Transmission Electron Microscope at 60 kV.

Biodissolution analyses

Time-resolved dissolution measurements were made on 8.7 μg/mL MnO₂ nanosheets and microparticles in phagolysosomal simulant fluid (PSF) (pH 4.76) with and without 1 mM cysteine (Supplemental Table S4). A 5:1 molar equivalence ratio between reductant (cysteine) and MnO₂ (requiring two electrons per mole for reduction from Mn(IV) to Mn(II)) was utilized to ensure that the reductant was not the limiting reagent in the experiment. The complete formulation of PSF is listed in Table S4. The effects of pH on MnO₂ nanosheet dissolution were determined using 2% and 70% nitric acid solutions and a series of buffer solutions ranging from pH 3–8 (pH 3–4, potassium phthalate buffers and pH 6–8, phosphate based buffers). Dissolution was monitored by time-resolved measurement of total soluble Mn in subsamples following removal of nanosheets or particles by ultrafiltration. Solids were removed using a 20 nm Anatop filter and total dissolved Mn concentration in the filtrate was determined by a Thermo Scientific iCAPtm 7400 ICP-OES.

Cytotoxicity

Cytotoxicity was determined after exposure of fish gill cells to MnCl₂, MnO₂ nanosheets or microparticles (5–100 ppm Mn ion; 8.7–158.2 μg/mL MnO₂) for 24, 48 or 72 h. At harvest, cells were washed with PBS and stained with a final concentration of 2 uM calcein AM and 4 uM ethidium homodimer 1 in phenol free L15. After 10 min incubation, cells were imaged with an Olympus IX81 microscope at 10x. Live (green fluorescence) and dead (red fluorescence) cells were quantified using Cell Profiler.

Mitochondrial shape and membrane potential

MitoTracker Deep Red FM was used to detect alterations in mitochondrial shape following exposure to MnCl₂, MnO₂ nanosheets or microparticles (1–20 ppm Mn ion; 1.6–31.6 μg/mL MnO₂). Mitochondria and nuclei were fluorescently stained with MitoTracker Deep Red FM (400 nM; Invitrogen M22426) and Hoechst 33342 (1:1000, Invitrogen H3570) respectively for 30 min at 19°C. Confocal images were obtained at 60x with an Opera Phenix High Content Screening System (Perkin Elmer) and presented as maximum intensity projections. Harmony software (Perkin Elmer) was used to identify individual mitochondria within cells, measure their width and length and calculate average mitochondrial ratios (width/length) per cell (Supplemental Table S5).

Tetramethylrhodamine, methyl ester (TMRM) was used to detect intact mitochondrial membrane potential in live cells. After exposure to Mn compounds, cells were washed with PBS and incubated in Image-iT TMRM reagent (5 nM, Invitrogen I34361) and Hoechst 33342 (1:10,000, Invitrogen H3570) in serum-free L15 media for 1 h. Cells remained in the TMRM/Hoechst media while widefield fluorescent images were obtained at 20x on an Olympus IX81 microscope. Image analysis was performed using Cell Profiler 3.1.8 software. Individual cells were identified by Hoechst 33342 staining (blue fluorescence) followed by propagation secondary object identification with TMRM staining (red fluorescence) and adaptive Otsu thresholding. TMRM integrated intensity was calculated per cell and averaged across 3 fields of view (100–250 cells each) per treatment.

Mitochondrial respiration assay and ATP production

Oxygen consumption rate (OCR), an indicator of mitochondrial respiration, was measured using a Seahorse XFe96 analyzer (Agilent)). RTgill-w1 cells were seeded into Seahorse 96 well plates and exposed to MnO₂ nanosheets (8.7–158.2 μg/mL), MnO₂ microparticles (158.2 μg/mL) or MnCl₂ (20 ppm) for 24 or 48 h. The Seahorse XF Cell Mito Stress Test (Agilent) was conducted according to manufacturer’s instructions. At harvest, cells were washed with PBS to remove MnO₂ nanosheets or microparticles. OCR values were measured at the start of the experiment (basal rate) and after the cells were challenged with 1.5 uM oligomycin, 1 uM FCCP and 0.5 uM rotenone/antimycin A sequentially. At the end of the assay, cells were stained with Hoechst 33342 and quantified on a Cytation 5 Cell Imaging Reader (BioTek). To quantify the number of cells per well, the number of cells within a single field of view was multiplied by 3.71 (based on the ratio of well dimension to the field of view dimension). Oxygen consumption rates were normalized to whole well cell counts. Experiments were run in triplicate with at least three technical replicates per condition. The following mitochondrial function parameters were calculated according to Seahorse protocols (Divakaruni et al. 2014):

1. Basal respiration = (3^rd basal measurement) – (non-mitochondrial oxygen consumption)

2. Maximal respiration = (maximum rate measurement after FCCP injection) – (non-mitochondrial oxygen consumption)

3. Spare respiratory capacity = (maximal respiration) – (basal respiration)

4. ATP production = (3^rd basal measurement) – (minimum rate measurement after oligomycin injection)

Statistics

Statistical analyses were performed with GraphPad Prism 8.4.2 software. Results were expressed as the mean ± standard error of at least three independent experiments. Cytotoxicity, mitochondrial ratios, TMRM intensity and oxygen consumption rates were analyzed using two-way ANOVAs followed by Dunnett’s multiple comparisons tests. Basal respiration, maximal respiration, spare respiratory capacity and ATP production rates were analyzed separately using one-way ANOVA tests followed by Dunnett’s multiple comparisons tests.

Results

MnO₂ nanosheet internalization

To begin our investigation into the interaction between MnO₂ nanosheets and rainbow trout (Oncorhynchus mykiss) gill epithelial cells, we first assessed cellular uptake of this 2D material. LysoTracker Red was utilized to visualize lysosomes and combined with confocal and DIC imaging to determine the localization of 2D MnO₂ nanosheets and MnO₂ microparticles within the cell. MnO₂ nanosheets co-localized with lysosomes were observed after both 24 and 48 h (Figure 1A). TEM imaging also revealed the presence of MnO₂ nanosheets in autophagosomes (Figure 1B). Interestingly, while MnO₂ microparticles also co-localized with lysosomes, no internalization within autophagosomes was observed.

Figure 1.

Lysosomal uptake of MnO₂ nanosheets and microparticles does not disrupt lysosomal membrane integrity. (A) Confocal images of the colocalization of MnO₂ nanosheets or microparticles with lysosomes after 24–48 h exposure. Lysosomes were stained with LysoTracker Red (pink fluorescence) and nuclei were stained with Hoechst (blue fluorescence). Arrows indicate examples of the colocalization of nanosheets or microparticles with lysosomes. Scale bar = 10 ?m. (B) TEM micrograph of a MnO₂ nanosheet within an autophagosome after 48 h exposure. Autophagosome is outlined by the pink box and the internalized MnO₂ nanosheet is indicated by the pink arrow. Scale bar = 400 nm. (C) Fluorescent images of cathepsin B localization in fish gill cells. Punctate cathepsin B localization (red fluorescence) was observed in untreated cells and following 48 h exposure to 50 ppm MnO₂ nanosheets, MnO₂ microparticles and soluble MnCl₂. Nuclei are stained with Hoechst (blue fluorescence).

Previous studies have shown that one-dimensional (fiber-like) high aspect ratio nanomaterials, such as multi-walled carbon nanotubes, can disrupt the lysosomal membrane, resulting in the release of cathepsin B and ultimately, cell death (Shi et al. 2011, Zhang et al. 2012). These MnO₂ nanosheets also have high aspect ratio, but are two-dimensional and it is currently unclear how 2D structures interact with lysosomal membranes. To determine whether 2D MnO₂ nanosheets disrupt the lysosomal membrane, we assessed intracellular localization of cathepsin B activity after a 48 h exposure (Figure 1C). Cathepsin B activity was localized in discrete puncta throughout the cytoplasm of untreated control cells suggesting that no disruption of the lysosomal membranes occurred. The same result was observed in cells exposed to MnO₂ nanosheets, microparticles and soluble MnCl₂. Together these results suggest that 2D MnO₂ nanosheets are internalized by fish gill cells prior to their dissolution and are localized within lysosomes and autophagosomes without disrupting lysosomal membrane integrity.

MnO₂ nanosheet dissolution

Our next question was whether the MnO₂ nanosheets could dissolve within the lysosomal environment. MnO₂ is highly insoluble in pure water (Wang et al. 2016), but has been reported to dissolve in some, but not all, biologically-relevant fluids to release the potentially toxic Mn²⁺ ion (Gray et al. 2020, Chen et al. 2014, Hao et al. 2016). At the receptor level, the biological response to MnO₂ nanosheets may thus reflect interactions with the solid nanosheet, or with the associated Mn²⁺ ions, or both. A detailed study of Mn solid-ion partitioning in biological fluid phases will therefore be useful to help interpret additional cellular response endpoints.

To address this we quantified the dissolution of MnO₂ nanosheets and microparticles in buffers that mimic the lysosomal environment (Figure 2). Lysosomal dissolution may be acid mediated (pH 4–5), or may be promoted by electron donors such as cysteine which have been reported to be present in lysosomes (Lloyd 1986, Pisoni et al. 1990). In phagolysosomal simulant fluid (PSF) supplemented with 1 mM cysteine (Figure 2A), both MnO₂ nanosheets and particles undergo rapid dissolution (< 3 h). In the absence of cysteine, a much slower dissolution occurs, which is < 20% complete after 48 h. This slow pathway may involve a kinetically-limited reaction with as-yet unidentified weaker electron donors in the PSF formulation, or may involve proton mediation and thus be intrinsic to the low-pH of the lysosomal environment. To explore this further, we conducted experiments in nitric acid solutions (Figure 2B) and in simple buffers spanning pH 3 to 8 (Figure 2C). A similar slow dissolution step is seen in these simple acidic solutions, along with pH dependent rates suggesting proton mediation (Figure 2D).

Figure 2.

Biochemical dissolution pathways and kinetics for MnO₂ materials. (A) Dissolution kinetics for both nanosheet and particulate forms in phagolysosomal simulant fluid (PSF) with and without 1 mM cysteine as a reductant; (B) Chemical dissolution rates for MnO₂ nanosheets in pure nitric acid solutions, room temperature; (C) pH-dependent chemical dissolution rates for MnO₂ nanosheets in series of buffer solutions (pH 3–4: KHP; pH 6–8: phosphate) at room temperature. (D) Two proposed pathways for degradative dissolution of MnO₂ materials in biological systems, consistent with literature and with data in panels A-C.

MnO₂ nanosheet cytotoxicity

The cytotoxicity of 2D MnO₂ nanosheets was then established using calcein AM and ethidium homodimer 1 stains for live and dead cells, respectively (Figure 3). Significant cell death was observed following 72 h exposure to 100 ppm MnO₂ nanosheets (158.2 μg/mL) (Figure 3A) while no significant cell death was observed following MnO₂ microparticle exposure (Figure 3B). Exposure to soluble MnCl₂ (100 ppm Mn) induced a time dependent decrease in cell survival, reducing cell viability to 49.4% after 24 h, 8.6% after 48 h and 3.2% after 72 h (Figure 3C).

Figure 3.

MnO₂ nanosheets induce cytotoxicity after prolonged exposure. (A) MnO₂ nanosheets (100 ppm) induce a significant decrease in cell survival after 72 h exposure (**p<0.005; n=4). (B) MnO₂ microparticles do not induce cytotoxicity through 72 h of exposure (n=4–8). (C) Soluble MnCl₂ (100 ppm) induces a significant decrease in cell survival in a time dependent manner. (****p<0.0001; n=4) (D) Representative images of live (green fluorescence) and dead (red fluorescence) cells after 72 h exposure to MnO₂ nanosheets, MnO₂ microparticles or soluble MnCl₂.

Impact of MnO₂ nanosheets on mitochondria

Previous studies have shown that excess intracellular Mn impairs mitochondrial function in both humans and aquatic species (Chen 2018, Crawford et al. 2011, Harischandra et al. 2019). Fish gills are highly metabolically active organs that are responsible for gas exchange as well as ionic and osmotic regulation (Evans et al. 2005). Impairment of mitochondrial function in gill cells could decrease ATP production, impairing ionic homeostasis of the gill tissue and subsequently, fish health and survival (Dawson et al. 2020). Here we investigated the effect of 2D MnO₂ nanosheets on mitochondria through three measures of mitochondrial integrity: mitochondrial shape, mitochondrial membrane potential and mitochondrial respiration rate. In order to ensure that any mitochondrial defects observed were not a consequence of nanosheet-induced cytotoxicity, the following experiments were conducted at sub-cytotoxic concentrations (1–20 ppm Mn, Figure 3).

Mitochondria can undergo alterations in morphology through fission, fusion, biogenesis and mitophagy in response to various cellular stresses (Kasahara and Scorrano 2014). Elongated mitochondria, resulting from mitochondrial fusion are crucial to respiratory active cells while round, fragmented mitochondria, resulting from mitochondrial fission, are frequently found in resting cells and terminally damaged cells (Westermann 2012). While conducing TEM imaging of fish gill cells treated with 2D MnO₂ nanosheets, we observed mitochondria with a “pinched” morphology after both 24 and 48 h exposures (Figure 4A), suggestive of mitochondrial fission. While untreated cells contained bean-shaped mitochondria with clearly defined membranes, cells exposed to 2D MnO₂ nanosheets exhibited mitochondria with indistinct membranes after 24 h and small, round mitochondria after 48 h (Figure 4A). We then quantified this alteration in mitochondrial morphology using confocal, single-cell imaging of fluorescently labelled mitochondria and automated mitochondrial identification, segmentation and analyses to determine the mean mitochondrial width/length ratio per cell. A concentration dependent increase in mitochondrial ratio was observed 48 h after exposure to 2D MnO₂ nanosheets (Figures 4B, 4C). No alteration of mitochondrial ratio was observed following exposure to MnO₂ microparticles or soluble MnCl₂.

Figure 4.

MnO₂ nanosheets induce alterations to mitochondrial morphology. (A) TEM micrographs of normally shaped mitochondria (blue arrows) in untreated cells, “pinched” mitochondria observed after 24 h and 48 h exposures to MnO₂ nanosheets (yellow arrows) and small, round mitochondria observed after 48 h exposure (orange arrows). Scale bar = 600 nm. (B) 48 h exposure to MnO₂ nanosheets, but not MnO₂ microparticles or soluble MnCl₂, induced a statistically significant increase in mitochondrial ratio (width/length) (*p<0.05; n=3). (C) Representative images of mitochondria staining with Mitotracker red after 48 h exposures. Identified mitochondria were used to quantify mitochondrial width and length with Harmony software. Scale bar = 10 ?m.

Mitochondrial membrane potential (MMP) is maintained by proton pumps in the mitochondrial membrane and is essential for ATP production through oxidative phosphorylation and mitochondrial homeostasis (Zorova et al. 2018). Mitochondrial membrane potential of RTgill-W1 cells was assessed using fluorescent imaging of the MMP indicator dye TMRM after 24 and 48 h MnO₂ nanosheet or microparticle exposures (Figure 5). Exposure to MnO₂ nanosheets induced a significant, concentration dependent decrease in TMRM fluorescence intensity, indicating a decrease in MMP, after 48 h (Figure 5B). No significant decrease in TMRM intensity was observed after 24 or 48 h exposure to either MnO₂ microparticles or soluble MnCl₂.

Figure 5.

MnO₂ nanosheets disrupt mitochondrial membrane potential. (A) Mitochondrial membrane potential, indicated by mitochondrial TMRM intensity, is not disrupted after 24 h exposure to MnO₂ nanosheets, microparticles or soluble MnCl₂. (B) Only MnO2 nanosheets induce a significant, dose-dependent decrease in mitochondrial membrane potential. *p<0.05; ***p<0.001; ****p<0.0001 (C) Representative images of TMRM fluorescent dye in untreated fish gill cells and after 48 h exposure to 20 ppm MnO₂ nanosheets. Nuclei = blue fluorescence, TMRM = pink fluorescence, scale bar = 10 ?m.

To assess if changes in mitochondrial morphology and membrane potential impacted mitochondrial function we measured mitochondrial respiration in live fish gill cells using the Seahorse Mito Stress assay. We assessed the oxygen consumption rate (OCR), a measure of cellular respiration, in cells exposed to 2D MnO₂ nanosheets, MnO₂ microparticles and soluble MnCl₂ for 24 or 48 h (Figures 6 and 7). MnO₂ nanosheets induced a significant decrease in basal respiration rates after 48 h of exposure (Figures 6D and 7D). This was a significant change from the 24 h MnO₂ nanosheet exposure, which did not alter basal respiration at all (Figures 6B and 7A).

Figure 6.

MnO₂ nanosheets inhibit mitochondrial respiration of fish gill cells. (A) Schematic of the Seahorse XF Cell Mitochondrial Stress Test Data Profile indicating the injection times of the four drugs that challenge mitochondrial function and the mitochondrial parameters tested. (B,C) 24 h exposure to 20 ppm MnO₂ nanosheets, MnO₂ microparticles and soluble MnCl₂ inhibits the maximal respiration rate of fish gill cells. (D,E) After 48 h exposure, only MnO₂ nanosheets inhibited the basal respiration rate (20 ppm) and the maximal respiration rate (10 and 20 ppm) of fish gill cells (*p<0.05; **p<0.005; ***p<0.001; n=3).

Figure 7.

MnO₂ nanosheets inhibit key mitochondrial respiration parameters in fish gill cells. (A-C) MnO₂ nanosheets (5–20 ppm), MnO₂ microparticles (20 ppm) and soluble MnCl₂ (20 ppm) inhibit maximal respiration and the spare respiratory capacity of mitochondria after 24 h exposures. (D-F) Only MnO₂ nanosheets inhibit the rate of basal respiration, maximal respiration and the spare respiratory capacity of mitochondria (10–20 ppm) after 48 h exposures (*p<0.05; **p<0.005; ***p<0.001; n=3).

After 24 hours of exposure MnO₂ nanosheets induced a significant, concentration dependent decrease in maximal respiration and spare respiratory capacity (Figures 6B, 7B and 7C). Interestingly, 24 h of exposure to MnO₂ microparticles and soluble MnCl₂ induced comparable decreases in maximal respiration and spare respiratory capacity (at 20 ppm) (Figures 6C, 7B and 7C). However, this pattern of inhibition was reversed after 48 h of exposure. At 48 h only the MnO₂ nanosheets induced significant, sustained decreases in maximal respiration and spare respiratory capacity (Figures 6D, 7E and 7F). Maximal respiration and spare respiratory capacity of cells exposed to MnO₂ microparticles or soluble MnCl₂ for 48 h returned to control levels.

One major consequence of impaired mitochondrial respiration is inhibition of ATP production. Mitochondrial respiration is a major source of ATP production and contributes most of the cell’s energy (Attene-Ramos et al. 2015, Bonora et al. 2012). The Seahorse Mito Stress assay was utilized to calculate mitochondrial ATP production after exposure to MnO₂ nanosheets, MnO₂ microparticles or soluble MnCl₂ (Figure 8). Mitochondrial ATP production was significantly inhibited after 48 h exposure to MnO₂ nanosheets (Figure 8B). This inhibition of ATP production occurred in a concentration dependent manner. No effect on ATP production was observed following exposure to MnO₂ microparticles or soluble MnCl₂. Broader impacts of these functional changes are discussed next.

Figure 8.

MnO₂ nanosheets inhibit ATP production. (A,B) The rate of mitochondrial ATP production is inhibited by MnO₂ nanosheets (10–20 ppm), but not MnO₂ microparticles or soluble MnCl₂, after 48 h exposure (*p<0.05; **p<0.005; n=3).

Discussion

The numerous uses of 2D MnO₂ nanosheets range from industrial to biomedical applications, making the environmental release of these materials inevitable. Previous studies showed that MnO₂ nanosheets display complex biochemical reactivity that has important implications for safety assessments (Gray et al. 2020). Therefore, it is crucial to understand how these unique nanosheets interact with target cells of an environmental indicator species, such as rainbow trout.

This study showed colocalization of the MnO₂ nanosheets with lysosomes, suggesting cellular uptake through the endolysosomal pathway. Our data are supported by a previous study that observed intracellular MnO₂ nanosheets within membrane-bound cytoplasmic vesicles (Gray et al. 2020). Our TEM imaging also revealed the presence of the MnO₂ nanosheets within autophagosomes. Autophagy often involves sequestration of foreign material by the lysosome and has been shown to mediate the storage of nano-sized particles (Popp and Segatori 2015). This may provide a mechanism for cellular sequestration of MnO₂ nanosheets that are slow to dissolve.

MnO₂ dissolution typically involves conversion of Mn(IV) to Mn(II) and is thus a reductive process initiated by electron donors (Wang et al. 2016), which may include glutathione (Hao et al. 2016, Gray et al. 2020, Yuan et al. 2018), cysteine (Gray et al. 2020), chloride (Andrani and Khan 2005), ascorbic acid (Gray et al. 2020) and peroxide (Fan et al. 2015). A recent study reported rapid (< 1 hr) dissolution in the presence of cysteine, glutathione, or ascorbic acid, and a much slower (> 72 hr), dissolution in some cell culture media, which was attributed to the presence of weaker reducing agents in the complex media formulation (Gray et al. 2020). Dissolution in cell culture media studied was sufficiently slow to allow intact nanosheets to reach cell surfaces during in vitro exposure and become internalized as intact solids (Gray et al. 2020). This same behavior is confirmed in the present study by images of intracellular MnO₂ nanosheets and microparticles inside lysosomes without any impact on lysosomal membrane integrity. It is therefore important to assess potential MnO₂ dissolution in the acidic conditions of the lysosomal compartment.

Chen et al. (2014) previously reported “break-up” of MnO₂ nanosheets in low pH environments, but the fluid phases often contained other biomolecules that may include electron donor(s). Surprisingly, our data show MnO₂ dissolution in the absence of an obvious electron donor in simple acidic solutions. Dissolution in HCl solutions has been reported, where Cl⁻ was suggested as the electron donating species (Adrani and Khan 2005). Outside of the biological and environmental literature, MnO₂ is reported to dissolve in strong acids including sulfuric acid solutions (Godunov et al. 2012, 2017) and in phosphoric acid (Islam and Rahman 2013). The sulfuric acid reaction evolves O₂ gas (Godunov et al. 2017) according to:

$${\text{MnO}}_2+2{\text{H}}^{+}\Rightarrow {\text{Mn}}^{2+}+{\text{H}}_2\text{O}+\mathrm{½}{\text{O}}_2\left(\text{evolved gas}\right)$$

This route does not require any external electron donor because it is a disproportionation in which Mn(IV) is reduced while lattice oxygen, O, is oxidized to O₂(g). These literature reports, together with the new data in this study, suggest that H⁺-mediated disproportionation of MnO₂ nanosheets can occur at lysosomal pH values, even in the absence of biological antioxidants or other external electron donors. Although this biodissolution by disproportionation under mildly acidic conditions is slow (~ 2 days) relative to the reductive route (~ minutes), it can occur over the time scales of in vitro cell experiments or particle clearance in vivo, and we thus propose this as a second distinct chemical pathway relevant for Mn²⁺ ion release in biological systems. In vivo, the combination of reducing agents, such as cysteine, and low pH in lysosomes makes it highly likely that MnO₂ undergoes some dissolution within the lysosomal compartment, releasing Mn²⁺ ions.

MnO₂ nanosheets induced mitochondrial dysfunction and cytotoxicity after prolonged exposures of 48 and 72 h, respectively. These delayed responses suggest that cellular uptake and intracellular dissolution are early steps leading to MnO₂ nanosheet toxicity. This is the first study to demonstrate that 2D MnO₂ nanosheets induce mitochondrial dysfunction. The mitochondrial toxicity observed occurred at sub-cytotoxic levels, suggesting it is not a result of mitochondrial damage that occurs during apoptosis or necrosis.

Our data show that 2D MnO₂ nanosheets induced mitochondrial morphological alterations and disrupted mitochondrial function after 24 h of exposure. TEM imaging revealed mitochondrial “pinching” indicative of mitochondrial fission. In addition, the highly sensitive Seahorse Mito Stress assay showed that MnO₂ nanosheets inhibited the cellular maximal respiration and spare respiratory capacity at this timepoint. Maximal respiration represents the maximal capacity of cellular substrate oxidation. This parameter defines a cell’s highest potential to respond to an increased energy demand (Divakaruni et al. 2014). The spare respiratory capacity is defined as the difference between basal and maximal respiration. This spare capacity provides the cell with the ability to respond to an increased ATP demand and withstand periods of stress (Divakaruni et al. 2014). Previous studies of mitochondrial function have shown that oxygen consumption and ATP synthesis by oxidative phosphorylation are affected by mitochondrial morphology dynamics. Mitochondrial fragmentation or fission has been shown to decrease mitochondrial respiratory capacity and impair ATP synthesis (Picard et al. 2013), consistent with our observations following prolonged exposure to MnO₂ nanosheets.

After 48 hours MnO₂ nanosheets induced a quantifiable, dose-dependent increase in mitochondrial width to length ratio, identified using quantitative, single-cell, fluorescent confocal imaging. These data indicate that exposure to MnO₂ nanosheets induces an increase in mitochondrial roundness, consistent with our morphological observations of mitochondrial fission. This conclusion is further supported by an observed loss of mitochondrial membrane potential following 48 h exposure to the MnO₂ nanosheets, a phenotype that has also been associated with mitochondrial fission (Suzuki et al. 2018).

Fragmented mitochondria are indicative of low respiratory activity (Westermann 2012). After 48 h of exposure to MnO₂ nanosheets an additional indicator of mitochondrial dysfunction related to respiratory activity was observed. Basal respiration rate, which represents the respiration used to drive ATP synthesis and that which is associated with proton leak (Divakaruni et al. 2014) was significantly decreased. In addition, maximal respiration and spare respiratory capacity remained suppressed in parallel with diminished ATP production rate following exposure to sub-cytotoxic doses of MnO₂ nanosheets.

The gills are highly metabolically active organs that regulate gas exchange and ionic homeostasis (Dawson et al. 2020). Mitochondrial activity is important for providing adequate energy in the form of ATP at basal levels and in response to stress. The ability to respond to external stress, such as a predator or alterations in salinity or water temperature, are crucial for fish survival. Impairment of mitochondrial function, particularly inhibition of basal mitochondrial respiration, maximal respiration and spare respiratory capacity, reflects severe damage to this vital organ. Prolonged gill dysfunction would be lethal to the individual. Our observations suggest that MnO₂ nanosheet exposure would severely impair mitochondrial function within the gill tissue, which would likely result in fish death.

Interestingly, the observed toxicity of the 2D nanosheets was significantly more rapid and severe than that of MnO₂ microparticles. The MnO₂ microparticles did not induce alterations in mitochondrial morphology or mitochondrial membrane potential. While MnO₂ microparticles decreased maximal respiration and spare respiratory capacity after 24 h, both parameters recovered by 48 h. This suggests MnO₂ microparticles induce a transient inhibition of mitochondrial function. This transient inhibition of mitochondrial respiration was also observed after exposure to soluble MnCl₂ for 24 h. The observed transient inhibition of mitochondrial respiration may be due to early high Mn levels, which decrease by 48 h. One limitation of this study is that we were not able to measure intracellular MnO₂ levels due to technical limitations.

Reviewing the full set of material characterization data for the nanosheets and microparticles (Tables S1a, S1b and Figure S1) show that the two samples differ in several important features and properties. Some combination of these properties is believed to be responsible for the higher toxicity of nanosheets relative to particles. These differences include surface area (~ 400 m²/g for nanosheets vs. 32 m²/g for particles), shape (atomically thin sheets vs. globular particles), crystal phase (δ-MnO₂ nanosheets vs. β-MnO₂ particles) and the presence of atomically thin, reactive edges in the case of the nanosheet material. Our results suggest that the higher toxicity of the nanosheets is related to their higher reactivity. High reactivity leads to faster dissolution, and may also enhance direct chemical interactions between material surfaces and cellular sub-structures. The fact that the nanosheet materials have a higher reactivity can be seen in Figure 1a, though this plot is not scaled optimally to feature the difference. Gray et al. (2020) shows a 60 min time frame for cysteine-mediated dissolution of the MnO₂ particles compared to almost instantaneous reaction between cysteine and the nanosheets. In addition, the 2D MnO₂ nanosheets depleted intracellular levels of the reducing agent glutathione to a greater extent than MnO₂ microparticles (Gray et al., 2020). We suggest that this enhanced reactivity, which is consistent with the higher physical surface area of the nanosheets and their active edges, is likely responsible for the enhanced biological response observed in the present study.

A second possibility is that the 2D structure of the MnO₂ nanosheets results in high reactivity at their atomically thin edges. These highly reactive edges may interact with intracellular organelles, such as mitochondria. Interaction between the nanosheets and the outer mitochondrial membrane could disrupt mitochondrial membrane potential, resulting in mitochondrial toxicity. While this scenario is possible, we did not observe physical interaction between the MnO₂ nanosheets and the mitochondria.

Two previous studies generated manganese oxide nanoparticles using electric arc-generation as surrogates for welding fumes and demonstrated partial dissolution in artificial lysosomal fluid (Stebounova et al. 2018) and uptake by lung alveolar epithelial cells with production of ROS and cell death after 72 h (VanWinkle et al. 2009). These observations are consistent with the results of this investigation of fish gill cell toxicity induced by exposure to chemically-synthesized 2D MnO₂ nanosheets. Collectively, these studies support a mechanism for cellular uptake of Mn oxide nanoparticles or nanosheets into lysosomes resulting in mitochondrial toxicity.

The biodissolution of 2D nanomaterials is a critical part of their safety assessments (Gray at al. 2018, 2020). Gray et al. (2018) introduced a hazard screening framework for 2D materials based on their biodissolution, degradation pathway and the toxicity of the resulting degradation products. Previously 2D MnO₂ nanosheets had been proposed to represent a “Class C” material within this framework, defined as “biosoluble with potentially hazardous dissolution products” (Gray et al. 2020). The data presented here support this categorization and further demonstrate the potential adverse biological and environmental impacts of these engineered 2D nanosheets.