Plastic particles adsorb to the roots of freshwater vascular plant Spirodela polyrhiza but do not impair growth

We investigated the effect of nano‐ and microplastics on the freshwater duckweed species Spirodela polyrhiza, a vascular plant. S. polyrhiza was exposed for 120 h to concentrations ranging from 102 to 106 particles·mL−1. We assessed effects on growth and chlorophyll production, and explored adsorption and absorption by way of confocal microscopy. For both nano‐ and microsized particles, no concentration‐dependent effects on growth were found (expressed as fresh weight, frond, and root sizes). In addition, chlorophyll concentrations were not significantly affected. Confocal microscopy indicated that nanosized plastic particles adsorbed externally to the duckweed, especially to the roots. Internalized plastic particles could not be detected. Nevertheless, given their important role in ecosystems as a food source for a range of organisms, the adsorption of plastic particles to S. polyrhiza roots as detected in this study can result in the transfer of plastic particles to diverse herbivorous species within the ecosystem.

Plastic debris frequently enters the natural environment, where it accumulates and acts as an environmentally persistent contaminant (Horton et al. 2017). Smaller particles such as nanoplastics (< 100 nm) and microplastics (> 100 and < 5 mm) (Koelmans et al. 2015) have gained considerable attention, because they are potentially bioavailable to many organisms (Wright et al. 2013). The environmental concentrations of such small plastic particles < 100 μm are not well known, because standardized procedures for collection, fractionation, characterization, and quantification are lacking, which results in underestimation especially for smaller particles sizes (Huvet et al. 2016;SAPEA 2019). Concentrations are expected to increase with decreases in particle size, and predicted concentrations of 50 nm particles range between 10 3 and 10 10 particlesÁmL −1 (Lenz et al. 2016). Accelerating production, deposition, and the bioinert character of plastics contribute to further growing environmental concentrations (Huvet et al. 2016;Horton et al. 2017;SAPEA 2019).
To date there are only a few studies that focus on the impact of plastic particles on primary producers (Yokota et al. 2017), of which only three focus on vascular plants. Kal cíková et al. (2017) reported that the exposure of duckweed (Lemna minor) to 30-600 μm plastic particles decreased root cell viability and growth. The two sediment-rooted macrophytes Myriophyllum spicatum and Elodea sp. exhibited reduced root to shoot ratios when exposed to 50-190 nm plastic particles, and M. spicatum also showed decreased shoot length for these nanoplastics and reduced main shoot length for 20-500 μm microplastics (van Weert et al. 2019). A study on cress (Lepidium sativum) found significant but transient effects of plastic particles on germination rates and root growth (Bosker et al. 2019).
The lack of research on vascular plants results in a major knowledge gap concerning the effects of plastic particles on ecosystem health (Eerkes-Medrano et al. 2015). For example, aquatic freshwater plants provide shelter for many organisms at higher trophic levels, and serve as food sources to herbivorous species in the water as well as in fringing ecosystems. To help address this knowledge gap, the objective of our study was to determine if plastic particles negatively impact the freshwater vascular plant Spirodela polyrhiza, a duckweed species. Therefore, we studied the effects of nanoplastics (50 nm) and microplastics (500 nm) on the growth of fronds, roots, and fresh weight, as well as the effects on chlorophyll content of S. polyrhiza. S. polyrhiza is a freshwater vascular plant at the base of aquatic food webs (Greenberg et al. 1992) and has commonly been used as an ecological indicator to assess the toxicity of substances because of its high sensitivity (Böcük et al. 2013). Additionally, to answer the question of potential transfer along the food web, we assessed adsorption and uptake of the nanoplastics.

Test materials
S. polyrhiza, a species of duckweed and a freshwater vascular plant, was obtained from a commercial source (MicroBioTests, Gent, Belgium). Spherical polystyrene fluorescent plastic particles (density 1.05 gÁcm −3 ) of 50 nm (red) and 500 nm (green) were used (Fluoro-Max Aqueous Fluorescent Particles; Thermo-Scientific, Waltham, MA, U.S.A.). To remove surfactants, plastic particles were cleaned prior to usage (see Supplementary Information).

Experimental design
Prior to the toxicity assessment, turions were germinated in a 48-well test plate with 1 mL of Steinberg growth medium for 72 h at 25 C with 6000 lux top illumination in an incubator (IPP110, Memmert GmbH, Schwabach, Germany). At the start of the experiment, plants were randomly placed in a 48-well plate, containing 1 mL the assigned treatment (n = 8 replicates/treatment; control, 10 2 , 10 3 , 10 4 , 10 5 , and 10 6 particlesÁmL −1 ), and incubated for 120 h at conditions as previous described.

Endpoints assessed
Growth endpoints Growth was assessed by measuring fresh weight, frond area and root length at 0 and 120 h. Before determining fresh weight, plants were carefully patted using Kim-Wipes. Total number of fronds and frond areas were determined by taking vertical photographs of test wells (Nikon D3100; 18-55 mm lens; Nikon, Miniato, Japan). To determine the total number of roots and root length, a photograph was taken using a digital microscope (AnMo Electronics Corporation, New Taipei City, Taiwan). Images were used to determine frond area and root length with Fiji software (v. 2.00-rc-67/1.52c) (Schindelin et al. 2012), and total number of fronds and roots were counted.
Average specific growth rates in fresh weight, frond area, and root length were calculated based on OECD protocol 221 (OECD 2006): with μ i − j average specific growth rate, N i measurement of variable at t 0 , and N j measurement of variable at t 120 . Subsequently, percentage inhibition of growth rate was calculated relative to the control: with %I r percentage inhibition in average specific growth rate, μC mean value for μ in the control, μT mean value for μ in treatment group.

Chlorophyll content
The extraction of chlorophyll pigments was performed in dark rooms and samples were stored on ice during the operation in order to prevent the degradation of chlorophyll pigments, following established procedures (Porra and Thompson 1989). Fresh fronds with a weight of 0.03 g were transferred into a 1.5 mL Eppendorf together with 0.05 g of quartz sand and 100% methanol. The samples were Dovidat et al. Effect of plastic particles on duckweed homogenized for 1 min at 30 Hz (Retsch Mixer Mill MM220, Retsch, Haan, Germany) and centrifuged for 1 min at 13,200 rpm (Eppendorf MicroCentrifuge 5415 D, Eppendorf, Hamburg, Germany). Of the supernatant fraction, chlorophyll a (Chl a), chlorophyll b (Chl b), and total chlorophyll were determined at 120 h for control, 10 2 , 10 4 , and 10 6 particlesÁmL −1 , according to established procedures (Lichtenthaler 1998) (for more details see Supplementary Information).

Nanoplastic particle localization
A separate experiment was conducted to explore potential adsorption and internalization of plastic particles. Briefly, S. polyrhiza was exposed to 10 14 particlesÁmL −1 of 50 nm red fluorescent nanoplastics for 120 h under conditions as previously described. Plants were placed on a glass slide and imaged employing an inverted LSM 880 microscope (Zeiss, Oberkochem, Germany) equipped with EC Plan-Neofluar 10×/0.30 M27 objective. Plastic particles were excited with a 543 nm helium-neon laser and detected using a 620-700 BP filter. Transmitted light was detected in a separate channel. In order to distinguish potential adsorption and internalization of plastic particles, z-stacks were obtained comprising 2.27-μm thick optical slices. In order to obtain an overview along the entire root length, we applied the tile scan option of ZEN microscope software (Zeiss, Oberkochem, Germany), stitching eight acquired scans of 642.86 × 642.86 μm into an 8 × 1 panoramic tile. The software Fiji was used to process the images.

Statistical analysis
All data are recorded and deposited in Dryad (Dovidat 2019). Statistical analyses were performed using the RStudio software (v. 1.1.456). ANOVA was used to assess differences among treatments. Normality and homogeneity of the data was tested using Shapiro-Wilk and Levene's test, respectively. When assumptions failed, statistical analyses were continued due to the robustness of ANOVA, but results were interpreted with caution if p was close to alpha. Interaction effects between concentration and particle size were assessed using two-way ANOVA, and concentration-dependent effects using one-way ANOVA. The significance level (α) was set at 0.05. When statistically significant differences were detected, a Dunnett's post hoc test was conducted. All test statistics are provided in Table S1.

Growth
There were no statistically significant interaction effects between size and concentration of plastic particles affecting fresh weight, single largest frond area, total frond area, frond number, single longest root, total root length, or root number (Table 1). The observed percent inhibition of these growth endpoints was not concentration dependent (Table S1). Only the 50 nm plastic particles significantly inhibited the growth of the total frond area by 5.81% for concentrations of 10 4 and 10 5 particlesÁmL −1 (Table S2). However, the assumption of homogeneity of variance was violated, and as the p-value is close to 0.05, these results need to be interpreted with caution. For all other growth endpoints, differences in growth inhibition were observed, but these were not statistically significant and did not follow a dose-dependent pattern (Table S2).

Chlorophyll concentrations
There was no statistically significant interaction between size and particle exposure concentration when comparing different exposure treatments for any of the measured chlorophyll Table 1. The effect of 50 and 500 nm plastic particles fresh weight, fronds, and roots of Spirodela polyrhiza after 120 h of exposure (n = 8 AE SEM). Statistically significant differences in comparison to the control, which are determined using Dunnett's post hoc test, are indicated with *(0.01 < p < 0.05).  (Table 2). Differences in the measured chlorophyll concentrations between exposure treatments were small. Only the plants exposed to 50 nm particles in a treatment of 10 4 particlesÁmL −1 exhibited large, but nonstatistically significant reductions in Chl b concentration up to 35% (Table 2).

Nanoplastic particle localization
Confocal microscopy indicated that the 50 nm nanoplastics adsorb externally on to S. polyrhiza, as demonstrated by red fluorescence. Particle densities were higher on the root shafts and tips (Fig. 1a) than on the frond lower epidermis (Fig. 1b).
The fluorescence displays irregular patterns of larger sizes than the 50 nm nanoplastics, which suggests clustering of the particles. In orthogonal projections, nanoplastic particles were detected surrounding the entire roots surface (Fig. 1c). No internalized particles could be detected.

Discussion
Here, we provide results on the impact of nano-and microplastics on a vascular plant, an area of research that is     understudied. We investigated the question if plastic particles negatively impact the growth and chlorophyll concentrations of the freshwater vascular plant duckweed. Additionally, we examined uptake and adsorption to provide indications for potential trophic transfer. Our results indicate no significant adverse effects of nano-and microplastics on S. polyrhiza, even when exposed to high concentrations. The absence of effects on duckweed growth, as observed in the current study, differs from a study on a closely related species of lesser duckweed (L. minor), in which significant adverse effects on root growth, but no effects on frond growth (Kal cíková et al. 2017). Importantly, the plastic particles used by Kal cíková et al. (2017) were approximately 1000 times larger than the particles used in our study, and the exposure duration was 48 h longer.
Research on other organisms has found that toxicity is further complicated by plastic particles with modified shape or function (Dris et al. 2015). In another study in our laboratory using the same 50 and 500 nm particles, we found significant effects on root growth of cress (L. sativum), although these effects were short-lived and transient (Bosker et al. 2019 In order to compare our results with other studies, Table 3 provides a summary of available studies on the impact of plastic particles on primary producers (Table 3).
Research on algae has resulted in mixed outcomes, with several studies reporting no effects on the growth of algae (Davarpanah and Guilhermino 2015;Lagarde et al. 2016) while others observed significant growth inhibition (Besseling et al. 2014;Sjollema et al. 2016;Zhang et al. 2017) (Table 3). This demonstrates the heterogeneity of findings, limiting the ability to make generalizable conclusions (Burns and Boxall 2018).
Effects of plastic particles on photosynthesis are similarly equivocal (Table 3). For example, Kal cíková et al. (2017) and Bosker et al. (2019) conclude that the exposure to plastic particles does not negatively impact photosynthesis, supporting the findings of our study. However, several studies on algae detected reduced concentrations of photosynthetic pigments (Bhattacharya et al. 2010;Besseling et al. 2014;Zhang et al. 2017). Only Sjollema et al. (2016) reported no effects of plastic particles on the photosynthesis of algae (Table 3).
There is little evidence for plastic particle uptake by vascular plants (Ng et al. 2018), with only one study known to us that found accumulation on the root hairs of the cress L. sativum (Bosker et al. 2019). In research on algae, however, the particle sizes used range from 20 nm (Nolte et al. 2017) to 2000 nm (Long et al. 2015). These two studies found adsorption without negative impacts, such as external adsorption of plastic particles to the plant tissue of the microalga Pseudokirchneriella subcapitata (Nolte et al. 2017) and increased accumulations of microplastics in algae aggregates compared to background levels (Long et al. 2015). Bhattacharya et al. (2010) observed that adsorption of positively charged, 200-nm sized plastic particles to algae reduced photosynthesis due to the physical blockage of light. The confocal microscopy in this study indicates external attachment of 50 nm plastic particles to the root tips and shafts of S. polyrhiza, but this could not be related to adverse effects, which is potentially due to different mechanisms of photosynthetic pigment reduction between algae and vascular plants. A second explanation could be that the photosynthetic pigments are not located in the roots but in the fronds/leafs, and the particles mainly adsorbed to the roots. In addition, most studies conducted to date on plants and algae are short-term acute exposures, highlighting the need to investigate the impact of chronic exposure of nano-and microplastics on plants. Furthermore, we could have missed potentially internalized particles due to limited penetration of the fluorescence signal through the root tissue. Nevertheless, our observed adsorption of plastic particles to the plant is important as adsorption might still contribute to biomagnification along the food web (Nolte et al. 2017). In addition, transfer to herbivorous species (both aquatic as well as terrestrial species feeding on aquatic plants) can occur, as we demonstrated that particles can attach and hence concentrate around the roots of duckweed.
To conclude, here, we present novel research on the effects of plastic particles on a freshwater vascular plant, and the first study to include nanoplastics. The results indicate that plastic particles of 50 and 500 nm do not negatively affect the growth and chlorophyll production of S. polyrhiza. Fluorescent imaging suggests, however, that the 50 nm nanoplastics adsorb externally. This study contributes to our understanding on the effects of microplastics on plants, an area which is currently understudied (Burns and Boxall 2018).