Fate and effects of microplastic particles in a periphyton-grazer system ☆

In the aquatic environment, microplastic particles (MP) can accumulate in microbial communities that cover submerged substrata, i


Introduction
Microplastics (MP) are a remarkably diverse and widely distributed group of pollutants of global concern (SAPEA, 2019).In streams and rivers, these small sized plastic pieces (1-5000 μm) are common (McCormick et al., 2014).Particularly high accumulation of MP has been found in benthic zones, unveiling those areas as environmental sink for MP (Besseling et al., 2016;Mani et al., 2015).Their residence time in a stream depends on the hydrological conditionsit has recently been estimated that, under low flow conditions, downstream transport of MP is slow and can reach up to 7 years per km (Drummond et al., 2022), suggesting high persistence of MP in streams and rivers.
In the benthic zones, MP can get incorporated in the microbial biofilms growing on submerged substrates (Sgier et al., 2016;Merbt et al., 2022).Such biofilms are referred to as periphyton with substrates being cobbles, sediment and coarse organic materials (Lock et al., 1984;Battin et al., 2003).Periphyton is a complex mixture of bacteria, archaea, algae and fungi.It has been suggested to immobilize MP via attachment (Guasch et al., 2022;Kalčíková and Bundschuh, 2022;Merbt et al., 2022).Since periphyton represents the main food source for many aquatic organisms, especially scarpers and grazers, the incorporation of MP in the periphyton may increase MP bioavailability to benthic organisms.It is unknown, however, whether MP incorporated into periphyton are transferred to aquatic grazers and whether they can cause adverse effects in the organisms.
Gastropods (marine and freshwater) ingest MP in different sizes and shapes as has been shown in field and laboratory studies (Scherer et al., 2017).For example, plastic fibers and films were found in marine snails, Littorella littorea (Doyle et al., 2019) and Colus jeffreysianus (Courtene--Jones et al., 2017), and in freshwater snails, Lanistes varicus, Melanoides and Theodoxus fluviatilis (Akindele et al., 2019).Potential consequences of MP ingestion on gastropods were, however, rarely characterized.Few studies reported a lack of effects on apical endpoints, such as emergence behavior (Littorina littorea) (Doyle et al., 2020), and feeding, defecation and reproduction rate (Potampoyrgus antipodarum, Lymnaea stagnalis) (Imhof and Laforsch, 2016;Weber et al., 2021).Yet, Jeyavani et al. (2022) recently found strong adverse effects upon MP (polyethylene beads, 12-45 μm) ingestion on the molecular and physiological level, such as increased oxidative stress and severe damage of digestive gland cells, in Pomacea paludosa upon 28 days exposure.All available studies were carried out with the MP spiked into food or mixed into the sediment.Co-ingestion of MP and periphyton has not yet been assessed.
MP can undergo physical and chemical changes when passing through the digestive tract.For instance, MP fragmentation has been observed upon ingestion by the arctic grill (Dawson et al., 2018) as well as leaching of chemicals from digested MP (Schrank et al., 2019;Schür et al., 2019;Sun et al., 2021).However, in vitro digestion assays showed that the digestive processes did not alter the chemical constitution of polystyrene-MPs but formed a corona of macromolecules on the MP surface (Liu et al., 2020;Stock et al., 2020).This alteration might enhance attachment of microbiota (Rummel et al., 2021(Rummel et al., , 2017) ) and as well induce changes in buoyancy (Rummel et al., 2017).Whether a MP experiences changes in its surface characteristics upon passing through the digestive tract of grazers has not been analyzed yet.
Based on the lack of knowledge of MP-periphyton-grazer interactions, we here investigated the fate of MP in a periphyton-grazer system and tracked MP uptake and consequences on feeding, defecation and reproduction of a freshwater snail upon MP ingestion.We selected the freshwater snail, Physa acuta, as model organism because it is a cosmopolitan grazer, which represents a significant part of many fish and waterfowl diet.Effects on the embryonal development of snails was examined after parental exposure but as well upon direct MP exposure of egg clutches.We moreover provide first evidence of the alteration of MP surface characteristics upon passage through the digestive tract of snails.

Microplastic
A spherical MP particle type was purchased at Cospheric (USA) as a dry powder (Product line: FMR).The MP, a thermoset amino formaldehyde polymer, were labeled with a red fluorophore, had a size range of 1-4 μm in diameter and a density of 1.3 g mL − 1 (MP).
The MP size distribution was confirmed by light microscopy using a DMI 6000 B (Leica, Suppl.Fig. S1).While the exact chemical composition is proprietary to Cospheric, the company stated that these particles are highly cross linked, that no leftovers from polymerization have been reported and that the red fluorophore is embedded.Concerning the polymer composition, Gerdes et al. (2019) applied FTIR on particles from the same product line (FM, only with green fluorescent dye), yet the polymer composition remained unidentified.The stability of MP in different suspension agents was tested earlier by our group and results showed that an extract of extracellular polymeric substances (EPS) from periphyton as medium yielded the best dispersion in terms of % MP over time and was therefore used for all exposure experiments and 2) the MP dispersions in EPS were stable for 75 min (Merbt et al., 2022).

Snail culture
Adult individuals of the freshwater snail, Physa acuta (Draparnaud, 1805), a pulmonate hermaphrodite gastropod, were collected from the indoor aquaria in the botanical garden of Zürich (Switzerland).Snails were transferred to indoor aquaria filled with 40 L treated tap water (300-350 μS cm − 1 , UVC treated and filtered through active carbon filter) and kept under constant aeration, and ambient light and temperature conditions (20-22 • C).Snails were fed with a mixture of fish food (TetraMin, Tetra, Blacksburg, VA) and fresh organic lettuce.

Design of the exposure experiments
One embryonal development test and three independent feeding experiments were carried out (Table 1).The embryonal development test aimed to examine whether direct MP exposure of egg clutches alter the embryonal development and hatching success of snail embryos.
The feeding experiments aimed to test whether the snails ingested MP embedded in periphyton and if so, if there were adverse effects on the snails' fitness.The effects on feeding rate, defecation rate, reproductive output, and on the embryonal development of snail embryos upon MP ingestion were recorded.Moreover, in a more detailed analysis, carried out exclusively in Exp. 2 (Table 1), the effects of the passage through the digestive tract on MP surface optical properties were quantified using flow cytometry.

Embryonal development test
To test if MP affect the embryonal development and hatching success of Physa acuta embryos, egg clutches were exposed to MP until hatch.To do so, snails were collected from the aquarium and placed into smaller glass containers (21 x 15 × 7 cm) filled with 300 mL artificial fresh water medium (PERIQUIL, composition see Stewart et al., 2015).PERIQUIL was selected since it mimics natural fresh waters, enhances the colloidal stability of the particles and allows clean (i.e., free of contamination) and standardized exposure conditions.Snails were held under ambient light and temperature conditions and the microcosms were inspected daily for fresh egg clutches.A total of 62 egg clutches with 20.8 ± 1.6 (SE, standard error) eggs per clutch (<24 h old) were immediately transferred into six well plates; one clutch per well (Exp.1, Table 1).The wells were filled with 10 mL PERIQUIL.For MP treatment, MP were suspended in an extract of extracellular polymeric substances (EPS, see below) from the periphyton to a concentration of 3.55 mg mL − 1 .A volume of 50 μL of this MP suspension was added to reach a final MP exposure concentration of 17.8 μg mL − 1 .For control treatment, 50 μL of EPS were added to the wells.The development of the snail embryos was examined every second day for abnormal growth, morphological deformations and decomposition, compared to control embryos at the same developmental stage (Musee et al., 2010).Hatching success was calculated over 28 days.Egg clutches were visualized using fluorescence microscopy (DMI 6000 B, Leica) in order to assess attachment to or internalization of MP into the egg clutches.

Feeding trials 2.5.1. Periphyton growth and extraction of extracellular polymeric substances (EPS)
Periphyton was needed for 1) the extraction of EPS to disperse MP and 2) to inoculate the glass containers that were used for the feeding trial.As previously described (Merbt et al., 2022), periphyton was grown from natural inoculum in flow-through indoor channels (880 × 110 mm, Plexiglas) on glass slides (76 × 26 mm, Schott) as previously described (Merbt et al., 2022).After 21 days, colonized glass slides were sampled from the channels and periphyton was scraped off.
For the extraction of EPS, freshly collected periphyton was suspended in 2 mM sodium hydrogen carbonate (Merck) and extracted as previously described (Merbt et al., 2022).
For the inoculation of the glass containers of the feeding experiment, the freshly collected periphyton was suspended in PERIQUIL to an optical density (OD) of 4 at a wavelength of λ = 645 nm (Cary 100 Thermo Fisher spectrophotometer, Kontron Instruments, Basel, Switzerland).

Experimental setup
The periphyton suspension obtained above was divided to serve as inoculum for the MP treatments during the feeding experiment.Treatments were control (no MP addition) or MP-periphyton.For controlperiphyton, 100 mL of periphyton suspension together with 10 mL of EPS were transferred to an Erlenmeyer flask.For MP-periphyton, 100 mL periphyton suspension and 10 mL of MP suspension (0.1 mg mL − 1 in EPS) were transferred to the Erlenmeyer flask.Both, MP and control periphyton were stirred slowly during 5 min using a magnetic stirrer.
Once the periphyton suspensions for inoculations were established, 24 glass microcosms (21 x 15 × 7 cm) were filled with 200 mL PERI-QUIL.Eight replicates per treatment were inoculated with 10 mL of the corresponding inoculation mixture (i.e.MP and control, Exp.2-4, Table 1) leading to a final MP concentration in the microcosms of 5 μg mL − 1 , corresponding to 2,5 × 10 − 3 m 2 MP per microcosm.
Microcosms were placed on three-dimensional orbital shakers rotating at 20 rpm with a vertical angel of 5 • (Edmund Bühler GmbH, Germany) at 18 • C and with a dark/light photoperiod for 12/12h under an LED light source (PAR).Similar to previous studies (Gil et al., 2015a), to maintain nutrient levels in the microcosms, the medium was fully changed every 3-4 days.After 14 days, the microcosms were completely covered with periphyton and were used in the feeding experiments.
Before starting the trial, adult snails with an average shell length of 7.8 ± 0.54 (SE) mm were placed in PERIQUIL during 24 h to allow acclimation.Afterwards, five snails per microcosm were placed in four out of the eight microcosms per treatment, and the other four microcosms per treatment were kept as controls without grazers.Snails were kept in the microcosms for 45 h, 152 h, and 88 h in Exp. 2, 3 and 4, respectively (Table 1).This was the period needed to consume approximately 70 % of the periphyton coverage in the microcosms (Supplemental Fig. S2), which was needed to calculate apical endpoints (see below).
Microcosms were inspected daily for new egg clutches and dead snails, which were removed.The experiment was stopped by removing the snails from the microcosms.Snails were briefly dipped in fresh PERIQUIL to wash off MP that could be loosely attached to the snails and placed into a depuration bath for 24 h.After the depuration time, snails were anesthetized in a 30 % MgCl 2 for 30 s (Musee et al., 2010), soft body tissue was separated from the shell, frozen at − 80 • C, and lyophilized in Lyovac GT2 (Leybold Heraeus) for 72 h.
The MP were quantified in periphyton before and after feeding, in the feces and in the soft body tissue of the snail using a Neubaur chamber and flowcytometry (see section 2.5.4 and 2.5.5).Moreover, in Exp.4, the hatching success of snail embryos was estimated (Table 1).

Biological parameter
Defecation rate: For calculation of the defecation rate, the fecal pellets produced during the feeding experiments were collected every second day of exposure and after the depuration phase by using a pipette, transferred into a pre-weighted falcon tube and centrifuged for 10 min at 1880×g.After discarding the supernatant, the fecal pellets were frozen at − 80 • C, lyophilized for 72 h and weighted.The dry weight was then determined as difference between lyophilized feces pellets and empty falcon tube.Defecation rate was expressed as produced fecal biomass per exposure and depuration time (mg per h).
Feeding rate: periphyton was scraped off from the bottom and the walls of the microcosms and was suspended in PERIQUIL.To obtain periphyton dry mass, 3 mL aliquots of the suspensions were filtered through pre-weighed glass fiber filters (GF/F, 25 mm diameter, 0.7 μm average pore size; Whatman Ltd., UK), then dried at 60 • C for 24 h and weighted.Dry mass was calculated as the mass difference between dried and empty filters and scaled to the surface area of the microcosms.The consumed periphyton was calculated as the difference in biomass between microcosms with and without snails within a treatment.The feeding rate was expressed as the consumed periphyton dry mass per day.
Reproductive output: The number of clutches and number of eggs per clutch over the experimental time were monitored and expressed as number of clutches per snail.Hatching success of the snail embryos (parental MP exposure) was determined in Exp.3.To do so, the clutches were transferred to six well plates containing 10 mL PERIQUIL.Embryonal development of the eggs in each clutch was monitored every second day by light microscopy.Hatching success was expressed as the % of eggs that hatched until day 28.If an egg did not hatch after 28 d, it was considered damaged (Musee et al., 2010).

Quantification of MP
At the end of Exp.3, MP were quantified in periphyton, snail soft body tissues and feces.A volume of 10 μL of the periphyton and feces suspension were transferred to the Neubauer chamber.The freeze-dried extracts of snail soft body tissue were suspended in 1 mL Tween 80 (0.01% w/w) by vortexing and 1 min sonication (45 kHz 60W, VWR Ultrasonic Cleaner).A subsample of 10 μL were transferred to a Neubauer chamber for MP quantification.The abundance of MP in periphyton was expressed in relation to the microcosm surface area (MP per m − 2 ) and per g periphyton dry weight (MP per g).The abundance of MP in the tissue and feces of the snails was expressed per g dry weight.

Characterization of microbial community composition and MP characteristics by flow cytometry (FC) and visual stochastic network embedding (viSNE)
Photoautotrophic microorganisms phenotypic community composition of the periphyton, the microbial composition of feces and the optical properties of MP suspended in periphyton, tissue and feces were analyzed by FC combined with viSNE data analysis (Sgier et al., 2016(Sgier et al., , 2018;;Merbt et al., 2022).This was done with samples obtained from Exp.1.
For sample preparation and FC run conditions, please refer to Supplemental Information (Text S1).The resulting FC data were analyzed applying viSNE by using the bh-SNE version of SNE (Amir et al., 2013;Van Der Maaten, 2014).Results were displayed in viSNE maps, where events measured by FC with similar fluorescence and scattering properties are plotted close to each other forming clusters.The abundances of the events per cluster in each viSNE map was quantified following Sgier et al., (2016).To identify to which main microbial groups the clusters were similar to (i.e.cyanobacteria, diatoms, green algae and red algae), the clusters were compared to the optical properties of reference species (Sgier et al., 2016(Sgier et al., ,2018)).For more detail refer to Supplemental information (Text S1).

Data analysis
Data distribuition test: All data were tested for normal distribution and homogeneity of variance using Shapiro-Wilks and Levene's test, respectively, setting the threshold for the null-hypothesis to p ≤ 0.05.If normality requirements were not fulfilled, tests were repeated with the log-transformed data.If normality were not given with log-transformed data, a non-parametric test was applied.
Data analysis of embryonal development test: Differences between hatching success of exposed and unexposed snails were identified using Student's t-test.
Data analysis of feeding experiment: Taking all three feeding experiments together (Exp.2-4,Table 1), all biological parameters, except mortality, varied significantly among experiments (one way ANOVA, p < 0.05, Supplemental Table S2), but not among treatments (Students T Test, p > 0.05, Supplemental Table S2), suggesting that the variability between experiments was greater than the treatment effect.Therefore, after assuring normality requirements (Supplemental Table S3), differences between response variables upon MP exposure were analyzed separately for each experiment using either Student's T-test or Wilcoxon test for normal and not-normal distributed data respectively.
Data analysis of FC & visne results: To determine the effect of MP on periphyton phenotypic community composition and on feces, the abundances of events per cluster in the respective maps were compared using Permanova followed by Tukey Test with Holmes correlated pvalues for multiple comparison using the following code in R https:// github.com/pmartinezarbizu/pairwiseAdonis/blob/master/pairwiseAdonis/R/pairwise.adonis.R.

Altered embryonal development upon direct exposure to MP
The exposure to MP of Physa acuta egg clutches taken from nonexposed snails showed that MP attach to the chorion as visualized by light microscopy (Fig. 1).Internalization of the MP into the egg clutch and the eggs was not observed.Hatching success decreased from 54 ± 6.4 % in non-exposed clutches (control) to 30 ± 5.2% (SE) in MP exposed clutches (Fig. 1).Embryonal development was not affected by the MP exposure.

No changes in biomass accrual and photoautotrophic microorganism phenotypic community composition in periphyton upon MP exposure
Overlays with data from reference species indicated that the phenotypic algal community was diverse: representatives of red algae, cyanobacteria, diatoms and green algae were detected by using FC and viSNE data analysis (Suppl.Fig. S3).Quantitative analysis of the clusters, however, indicated no differences in periphyton community composition between control and MP periphyton (Suppl.Fig. S4).Along these lines, MP had no effect on biomass accrual of periphyton (p > 0,05, Students T test, Suppl.Fig. S5).
The feeding of the snails caused a significant decrease in periphyton biomass in Exp. 2 and Exp. 4 (Suppl.Table S4).In Exp. 3, mean periphyton biomass was reduced in both treatments, however, differences were not statistically significant (p > 0.05, Student's t-test, Suppl.Table S4).

Snails show no selective feeding strategy upon MP exposure
Quantification of MP by light microscopy in periphyton demonstrated that the abundance of MP per microcosm surface (m 2 ) decreased significantly (p ≤ 0.05, Wilcoxon test) upon grazing (Table 2).However, MP abundance per periphyton biomass (mg dry weight, Table 2) was similar before and after feeding, suggesting a non-selective feeding strategy of the snails, i.e. periphyton and MP were ingested at the same rate.

Changed microbial community composition of snail feces exposed to MP; MP can remain in snail tissue
MP were present in the snail feces produced during the exposure and depuration (Fig. 2) confirming their ingestion.MP abundance in the feces was analyzed with both, light microscopy and FC and viSNE (Table 2 and Supplemental Table S5).
The abundances of MP per dry weight periphyton doubled per dry weight feces suggesting an increase in concentration of MP in the feces (Table 2).Moreover, based on optical properties, the microbial community composition of the feces produced by snails feeding on control and MP containing periphyton, respectively, differed significantly (Fig. 3).This difference in composition of the feces disappeared during the depuration phase (Fig. 3), despite the fact that relative abundance of MP in the feces excreted during exposure and during depuration was similar (Supplemental Table S5).
Based on light microscopy, no MP were detected in snail tissue after depuration phase (Table 2) whereas FC and viSNE analysis allowed the detection of small amounts of MP (Supplemental Fig. S6).This highlights the higher sensitivity of FC and viSNE approach compared to light microscopy.

Decreased snail fitness upon MP exposure
Snail mortality, feeding rate and defecation rate were not significantly altered by any MP treatment (Table 3).In contrast, MP exposure had a significant effect on reproductive output.In particular, the number of clutches per snail was significantly lower in Exp. 2 and 4 upon MP exposure, while the number of eggs per clutch was not affected.In contrast, in Exp.3, this pattern was reversed with the number of clutches per snail not being affected and the number of eggs per clutch being reduced in presence MP.In Exp.4, in general the number of clutches per snail was low compared to the other experiments.The hatching success from these clutches (parental exposure MP) indicated no adverse effects (Table 3).

The peripyhton-grazer system changes optical properties of MP
The incorporation into periphyton as well as the passage through the digestive tract of the snails induced changes in the optical properties of the MP.This was indicated by the fact that in the viSNE maps from MP containing periphyton and feces, the MP were split in two clusters (MP-CL1 and MP-CL9, Fig. 4).Moreover, these two clusters differed from the cluster of the MP Starting suspension.The most prominent differences in fluorescence intensities between the MP starting suspension, the biological matrixes (periphyton and feces) and MP-CL1 and MP-CL9 were observed at wavelengths 450 and 575 nm.The MP starting suspension has low and high intensities at 450 and 575 nm, respectively.This specific pattern helps identifying MP clusters in the map.At 450 nm in particular, MP-CL1 and MP-CL9 cluster had higher fluorescence intensities than the biological matrixes and the starting suspension of MP (FL9, Fig. 4B).High fluorescence at 450 nm was found earlier to be typical for non-labeled MP (Sgier et al., 2016).In contrast, at 575 nm, the MP-CL1 and MP-CL9 clusters were higher than the biological matrix, but lower than the MP starting suspension (Fig. 4C).
Relative abundances of MP in the clusters differed between biological matrixes.In periphyton, 97.9 % of MP were clustered in MP-CL1, while in feces, MP were more distributed between CL1 and CL9 containing approx.80 % and 20 % of total MP, respectively (Supplemental Table 2 MP content in periphyton, feces and snail soft body tissue in microcosms with and without snails after 88 h of exposure (Exp.4) analyzed by using light microscopy.For statistical analysis, data were log-transformed to meet normality requirements as tested with Shapiro and Levens' test.Bold values indicate statistically significant differences between treatments with and without snails (p ≤ 0.05, Student T test, unequal variance).Dw: dry weight.AVG: average, SE: standard error, n = 4. Table S5).This difference indicates a shift in optical properties of MP upon digestion.Yet, no changes in forward and sideward scatter (FS and SS) were observed, suggesting no changes in size and surface of the MP during digestion (Supplemental Fig. S7).
In snail tissue, MP had lower fluorescence intensities at 450 and 575 nm compared to the original red MP in the starting suspension, similar to MP in periphyton and feces, (Supplemental Fig. S7).The overall relative abundance of MP in tissue was low, resulting in less than <0.5 % of the measured events.

Discussion
The tested spherical MP were incorporated into the periphyton, yet their presence had no significant impact on periphyton biomass accrual and photoautotrophic community composition.This suggests that the higher surface availability due to the presence of MP in the exposures did neither induce an accelerated growth of periphyton nor select for certain photoautotrophic species.This is in line with a previous study applying the same experimental set up, were also no increased periphyton biomass accrual and no changes in the photoautotrophic community composition were observed when MP exposure concentrations were 2.5 times lower (Merbt et al., 2022).Hence, periphyton structural properties might be rather independent of the presence of MP and are shaped by other factors, e.g. by nutrient concentrations.
Once introduced in the microcosms, snails readily started to graze on both, control and MP-containing periphyton.MP were ingested and excreted by the snails.To evaluate a possible selective feeding behavior of the snails, we compared the concentrations of MP with respect to surface area (m 2 ) and periphyton biomass (mg) before and after the feeding experiment.Selective feeding would have resulted in an altered ratio of MP to periphyton biomass at the end of the experiments, which was not the case.This suggests that the snails ingested both, peripyhton and MP at a similar rate without any specific preference.The same indiscriminate feeding activity was described earlier in exposure experiments where snails fed on periphyton loaded with different amounts of fine inorganic material (i.e.sediment, <62 μm in diameter, Broekhuizen et al., 2001).Moreover, the anatomy of Physa acuta allows MP ingestion.The headcapsula is big enough to ingest even 90 μm sized fluorescent polystyrene spheres (Scherer et al., 2017) and the radula consists of 30-40 V-shaped rows of approximately 120-160 comb-like teeth that seem to be designed for rather unspecific grazing (Wethington et al., 2009).Previous studies also demonstrated that MP were ingested by other aquatic organisms (e.g.Acartia longiremis, Calanus finmarchicus) where ingestion depended on MP size (polystyrene fragments) and life stage of the organisms (Vroom et al., 2017).Selective feeding has, however, been observed in a recent study showing that copepodes rejected 80 % of all offered MP types differing in size, shape, presence of biofilm, adsorbed pollutants and material properties suggesting taste discrimination (Xu et al., 2022).
After ingestion, snails barely retained the MP as only a scant number of MP were detected in the soft body after depuration of 24 h.This suggests that long-term retention of MP in snails is rather unlikely.These

Table 3
Mortality, feeding rate, defection rate, number of clutches per snail, number of eggs per clutch and hatching success (only determined in Exp. 4) in control and MP treatment (mean ± SE) from Exp. 2 (n = 8), Exp. 3 (n = 12) and Exp. 4 (n = 12).Data were log transformed to meet normality requirements, as tested previously using Shapiro-Wilk and Levene's test (results are represented in Supplemental Tables S2 and 3).For normal distributed data, statistical differences were tested with Student's t-test.For not normal data, statistical differences were tested with Wilcoxon test and results were marked with (*).Bold values indicate significant differences (p ≤ 0.05).Ns: no significant differences tested.SE: standard error.Original values are represented in Supplemental Table 6  Lymnaea stagnalis where most of the ingested MP (10-90 μm sized, spherical polystyrene beads) were already excreted after 24 h (Weber et al.., 2021).Interestingly however, low amounts of MP remained after depuration as shown by FC and viSNE (present study) or even increased in abundance after 24 h of feeding (Weber et al., 2021) in the snails soft body.This is contradictory to the proposal that MP egestion is size dependent with smaller MP (1 μm) being egested faster than bigger MP (90 m, μ Kinjo et al., 2019) and to findings from exposures with polystyrene nanoplastic (approximately 300 nm) that were completely cleared in Physa acuta after 24 h (Holzer et al., 2022).Weber et al. (2021) attributed the increase of smaller MP (5 and 10 μm) in snails after 24 h to a heterogeneous distribution of the tested MP sizes due to the adsorption to the food source (i.e.lettuce), which increases the probability of ingestion by the snails.In the present study, a re-ingestion of MP via the feeding on feces in the depuration phase cannot be ruled out and could have contributed to the very low amount found after depuration.The ingestion of MP reduced the fitness of the snails.In particular, snail mortality, feeding and defecation rate were not affected, but an alteration of the reproductive output was observed upon trophic transfer of MP.The lack of effects on the feeding rate can be due to the fact that snails were not able to increase ingestion rate or assimilation efficiency of food to compensate for the increase of inorganic content in the food source (Broekhuizen et al., 2001;Ryder, 1989;Holzer et al., 2022).Consequently, in longer exposures (weeks, months), a state of malnutrition possibly sets in.
Similarly, defecation rate was not affected by MP ingestion.Yet, the feces of snails feeding on MP containing periphyton was significantly different to feces from non-exposed snails.This suggests that the periphyton components were digested differently in MP and control treatment.It is not possible, however, to conclude whether the digestion was more or less complete in either treatment, but indicates that MP alter digestion activity.There was no direct effect on mortality within the short exposure time (42-152 h).Prolonged exposure could elucidate whether ingestion of periphyton mixed with MP results in underfeeding.
The observed reduced reproductive output suggests that MP in the food source pose a physiological stress for the snails and hence the snails induce energy allocation strategy, i.e. inverting more energy to maintain metabolism/survival instead of reproduction.Reduction of reproductive output is a stress coping strategy.In Oysters for example, a decreased reproduction and increased structural growth has been observed during 2 months of exposure to polystyrene MP in the size of 6 μm (Sussarellu et al., 2016).It is therefore interesting that similar responses (i.e.lower reproduction) were observed in the present study after only 48-152 h of exposure.
The observed effect of MP on snail reproductive output did not translate to malfunctioning of the embryogenesis (growth, pigmentation, possible deformations, hatching success) in the snail embryos originated from feeding experiments (maternal exposure).Embryotoxicity can be induced due to maternal transfer as previously observed for silver nanomaterial in Physa acuta egg clutches (Gil et al., 2015b).The absence of such effects in the present study suggests that MP do not undergo translocation.
In contrast, when egg clutches originating from non-exposed snails were directly exposed to MP, the hatching success of the embryos decreased significantly.Together, this suggests that the effect is rather triggered from the outside than passed on from one generation to the other.Based on analysis by light microscopy, MP were not internalized but were tightly attached to the chorion by embedding into the mucous layer of the outer side of the chorion.Therefore, the observed reduced hatching success of MP exposed snail embryos can either result from physical disruption of the egg clutch by MP, or a reduced oxygen intake due to coverage of the chorion by MP, as previously hypothesized in a polystyrene MP (6 μm) exposure experiment with embryos of the marine fish, medaka (Oryzias melastigma) (Chen et al., 2020).Alternatively, internalization of the fluorophore leached from the MP could possibly have affected hatching of the snails.It has indeed previously been shown that the leached fluorophore from the same MP may translocate from Daphnia magna gut into other parts of the body (Schür et al., 2019).We observed that, while light scattering characteristics (FS and SS) of MP remained the same, the fluorescence intensity at the characteristic wavelength of MP were lower in MP in periphyton, in the digestive tract and in the feces than in the starting MP suspension.This may indicate loss of fluorophore from the MP.However, a loss of fluorescence can also be due to a coating of the MP surface by macromolecules, since MP have been shown to acquire what has been coined an "eco-corona" of organic molecules (Galloway et al. 2017;Liu et al., 2020).As the incorporation into periphyton and the passage through the digestive tract of the snail differentially induced alterations of MP fluorescence, we suggest that the main driver of change in optical properties were absorbed organic molecules.In sum, our study showed reduced reproduction output adult snails and reduced hatching of snail embryos upon MP exposure.Unveiling the mechanism of action underlying those effects such as changes in hormonal activity, heatshock protein expression upon MP exposure is a new avenue to be investigated in the future, which was, however, beyond the scope of the current work.

Conclusion
Periphyton is a sink for MP, with unknown and probably highly variable residence time.This study showed that the presence of MP has a range of effects in an aquatic periphyton-grazer system.Besides impacting microbial community composition of periphyton (Merbt et al., 2022), the incorporation into periphyton increases the bioavailability of MP for aquatic grazers, which themselves are important prey for fish and invertebrates.Hence, the ingestion of MP by grazers can promote the MP transfer to higher trophic levels.
MP ingestion entailed lower reproductive output, which indicates a critical effect on overall fitness of the snails.Moreover, reproduction is crucial for survival of species, and the stability of snail populations might be put at risk due to lower reproductive output.It cannot be ruled out that chronic exposures, over several months, to MP in the food source would lead to an amplification of the effects observed in the present studies.For instance, reduced locomotion, fitness and reproductive output can be expected, which will have substantial effects on grazing activity.Consequences of grazing are increased spatial heterogeneity and changes in the community composition of the periphyton, which together can increase resilience of periphytic communities (Katano and Doi, 2019).Hence, a decrease in grazer population in streams may disturb aquatic ecosystems function.

Funding
The study was financially supported by the Velux foundation, project number 1039, Switzerland.Additional lab work was funded by Tailwind grant of Eawag Switzerland.Open access funding was provided by Eawag-Swiss Federal Institute of Aquatic Science And Technology.

Declaration of competing interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Stephanie Merbt reports financial support was provided by Velux Foundation.If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 1 .
Fig. 1.Impact of MP exposure on hatching success of embryos taken from unexposed snails.The MP resulted in significantly reduced hatching success (A) based on Student's t-test (p < 0.05).The control indicates the hatching success of non-exposed egg clutches.Whiskers indicate the standard error of the mean (n = 31 clutches).MP were not internalized into the egg clutches, representative image (B) and attached to the outer layer of the egg clutches as highlighted by the arrow in the enlargement.

Fig. 2 .Fig. 3 .
Fig. 2. Representative microscopy images of snail feces after feeding on periphyton containing MP (red colour).Arrows indicate the MP within the feces.Images were taken from samples resulting from Exp. 3. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4 .
Fig. 4. Fate of MP in periphyton and feces.Panel A shows an overlay of periphyton (black), feces (green) and feces after depuration (red) and starting MP suspension in EPS (blue, and highlighed with the red circle).Cluster highlighted in pink and with green circle (MP CL1) and yellow with yellow circle (MP CL9) represent MP as part of the periphyton and snail feces, respectively.Fluorescence intensities at 450 nm (B) and 575 nm (C) are shown in panel B and C, respectively, indicating different optical properties of MP-CL1 and MP-CL9 compared to MP starting suspension and biological matrixes.(For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) Writingreview & editing, Writingoriginal draft, Methodology, Investigation, Data curation, Conceptualization.Alexandra Kroll: Writingreview & editing, Software, Methodology, Data curation.Linn Sgier: Software, Methodology, Data curation.Ahmed Tlili: Writingreview & editing, Conceptualization.Kristin Schirmer: Writingreview & editing, Funding acquisition, Conceptualization.Renata Behra: Writingreview & editing, Supervision, Funding acquisition, Conceptualization.

Table 1
Experimental design.