Snail communities increase submerged macrophyte growth by grazing epiphytic algae and phytoplankton in a mesocosm experiment

Abstract The relationships between producers (e.g., macrophytes, phytoplankton and epiphytic algae) and snails play an important role in maintaining the function and stability of shallow ecosystems. Complex relationships exist among macrophytes, epiphytic algae, phytoplankton, and snails. We studied the effects of snail communities (consisting of Radix swinhoei, Hippeutis cantori, Bellamya aeruginosa, and Parafossarulus striatulus) on the biomass of phytoplankton and epiphytic algae as well as on the growth of three species of submerged macrophytes (Hydrilla verticillata, Vallisneria natans, and one exotic submerged plant, Elodea nuttallii) in a 90‐day outdoor mesocosm experiment conducted on the shore of subtropical Lake Liangzihu, China. A structural equation model showed that the snail communities affected the submerged macrophytes by grazing phytoplankton and epiphytic algae (reduction in phytoplankton Chl‐a and epiphytic algal abundance), enhancing the biomass of submerged macrophytes. Highly branched macrophytes with high surfaces and morphologies and many microhabitats supported the most snails and epiphytic algae (the biomass of the snail communities and epiphytic algae on H. verticillata was greater than that on V. natans), and snails preferred to feed on native plants. Competition drove the snails to change their grazing preferences to achieve coexistence.

Phytoplankton and epiphytic algae are the main primary producers that compete with macrophytes for light, nutrients, and space (Arthaud et al., 2012;Phillips et al., 2016;Song et al., 2017).
Phytoplankton and epiphytic algae are also the main food sources of aquatic animals in freshwater ecosystems, such as fish, shrimp, snails, oligochaetes, mayflies, and chironomids (Asch et al., 2019;Chen et al., 2020;Guasch et al., 2016). Epiphytic algae and phytoplankton are considered the key factors causing the transformation between clear and turbid states in shallow aquatic ecosystems (Phillips et al., 2016;Qin et al., 2013). In the turbid state, the establishment and growth of submerged macrophytes may be restricted due to light attenuation induced by high phytoplankton and epiphytic algae biomass (Arthaud et al., 2012;Hidding et al., 2016), while high grazing pressure from predators reduces the biomass of phytoplankton and epiphytic algae, which then increases light availability and promotes macrophyte growth in the clear water state (Hilt, 2015;Sánchez et al., 2010).
Freshwater snails filter feed on phytoplankton in the water, scrape organic detritus, and periphyton from surfaces and sometimes also feed on macrophytes (Cao et al., 2014;Li et al., 2009;Yang et al., 2020). Most of freshwater snails are scrapers, and others are collector-filterers (Mo et al., 2017). Scrapers consume mainly epiphytic algae, but their diet also includes detritus and aquatic plants (Li et al., 2009). Collector-filterers use gills to filter suspended algae from the water column (Yang et al., 2020). Snail-algae interactions may thus be of great importance for submerged macrophytes. The grazing of epiphytic algae and phytoplankton increases the growth rates of macrophytes, potentially by reducing competition for light and/or nutrients (Brönmark, 1989;Yang et al., 2020). The above phenomenon is called a snail-macrophyte mutualistic interaction (Carpenter & Lodge, 1986;Li et al., 2007). Macrophytes, however, are also grazed by snails, which may have a significant impact on macrophyte growth (Elger & Lemoine, 2005;Li et al., 2009;Xiong et al., 2008). For example, Radix swinhoei, a member of Lymnaeidae, not only scrapes organic detritus and periphyton from the surface but also feeds on macrophytes (Li et al., 2006). Therefore, the relationship between snails and macrophytes remains unclear. Snails also exhibit complex and plastic behaviors when coexisting with other snails (Lombardo & Cooke, 2004). Overlapping food sources of freshwater snails may lead to competition (Holomuzki & Hemphill, 1996), and changes in resource utilization by competing snail species may impact food web dynamics and community assembly (Estebenet et al., 2002). However, studies of interspecific interactions among freshwater snails are uncommon (Dubart et al., 2019;Turner et al., 2007).
The ecological mechanisms by which snail communities affect macrophyte growth, phytoplankton biomass, epiphytic algal communities, and nutrient cycling and transformation are unclear. We hypothesized that snail grazing on both epiphytic algae and phytoplankton can indirectly improve the growth of submerged macrophytes. We further hypothesized that competition drives snails to change their grazing preferences to achieve coexistence, which leads snail communities toward maximal resource utilization. To test our hypotheses, we conducted an outdoor mesocosm experiment to elucidate the effects of snail communities on aquatic ecosystems.
We studied the effects of snail communities (consisting of Radix swinhoei, Hippeutis cantori, Bellamya aeruginosa, and Parafossarulus striatulus) on the biomass of phytoplankton and epiphytic algae as well as on the growth of three species of submerged macrophytes (Hydrilla verticillata, Vallisneria natans and one exotic submerged plant, Elodea nuttallii) in a 90-day outdoor mesocosm experiment conducted on the shore of subtropical Lake Liangzihu, China.

| Experimental design
An outdoor mesocosm experiment was conducted at the National Field Station of the Freshwater Ecosystem of Liangzi Lake (hereinafter referred to as Liangzi Lake Station), Hubei Province, China. A two-way factorial experiment was carried out with three species of submerged macrophytes (Hydrilla verticillata, Vallisneria natans, or Elodea nuttallii, the three species macrophytes were planted in their respective experimental vessels) and two grazing treatments (four species snails present or snail absent), with six replicates for each treatment, resulting in a total of 36 aquariums. The study began on Thirty-six glass fibre-reinforced polymer (GFRP) aquariums (inner diameter: 40 cm, height: 70 cm, Figure S1 ) were placed on a cement platform (50 m long, 20 m wide, Figure S1). The sediment used in our experiment was collected from Liangzi Lake. To ensure homogeneity and remove benthic animals (especially snails) before the experiment began, the sediment was air dried under natural conditions, ground, sieved (0.6 mm mesh size), and mixed before being added to the aquarium. To each aquarium, we added 10 cm of sediment (nitrogen content: 0.56 ± 0.05 mg g −1 , phosphorus content: 1.63 ± 0.02 mg·g −1 , organic matter content: 0.068 ± 0.003 mg g −1 ; all values are means ± SD). We subsequently added 70 L of groundwater (total nitrogen (TN): 0.52 mg L −1 and total phosphorus (TP): 0.03 mg L −1 ).
Hydrilla verticillata and V. natans are the dominant macrophytes in Liangzi Lake (Wang et al., 2019;Xu et al., 2018), and E. nuttallii is an invasive species in China (Xiong et al., 2008). On August 21 2017, 72 specimens of the submerged macrophytes H. verticillata, V. natans, and E. nuttallii were collected from a homogeneous population in the nursery ponds of the Liangzi Lake Station. All plants were carefully washed to remove snail eggs and periphyton, and six shoots of each macrophyte species were planted in each aquarium.
On September 21, 2017, a large number of vigorous and sexually mature snails Radix swinhoei, Hippeutis cantori, Bellamya aeruginosa, and Parafossarulus striatulus were collected from the macrophyte plants growing in the nursery ponds of Liangzi Lake Station. The snails were kept without food for 24 h before being added to the aquarium (Xiong et al., 2008). Subsequently, we selected 360 individuals of four species snails of homogeneous size and age. Of these species, R. swinhoei and H. cantori are hermaphroditic and undergo allogeneic fertilization, while B. aeruginosa and P. striatulus are dioecious (Li et al., 2009). Therefore, the ratio of females to males that we selected for B. aeruginosa and P. striatulus in this study was 1 to 1. After the submerged plants had grown for over one month (on were added to each aquarium, which was then covered by a nylon net (1.0 mm mesh size) to prevent snail escape. The fresh mass of the snail species was as follows: R. swinhoei: 0.38 ± 0.04 g ind. −1 , H. cantori: 0.04 ± 0.01 g·ind. −1 , B. aeruginosa: 2.33 ± 0.15 g ind. −1 , and P. striatulus: 0.16 ± 0.01 g ind. −1 . The water level of the aquariums regularly topped up to the initial level with pure water during the experiment.

| Epiphytic algae
Fifty leaves of H. verticillata, 50 leaves of E. nuttallii, and five leaves of V. natans were carefully selected to ensure uniformity in growth state and size before placing each into a wide-mouth plastic bottle with 200 ml of pure water in the respective aquarium.
Periphyton were removed with a banister brush in water (Foerster & Schlichting, 1965) and preserved in a well-labeled plastic container, with 2 ml of Lugol's solution to fix them. The area of selected leaves was measured with an area meter (LI-3100C, LI-COR). The epiphytic algae sample was centrifuged at 1788.8 g for 10 min, and the supernatant was discarded. Then, the volume was adjusted to 30 ml and mixed. The number and species of epiphytic algae were counted using a counting plate at 400× under an optical microscope. For each sample, 50 microscopic fields of vision were examined and counted. (Effiong & Inyang, 2015;Hu & Wei, 2006;Qian et al., 2015). Species richness (S) of each sample was quantified as the number of species in the sample, and the abundance (N, cells) was the total number of individual quantities and calculated using the following formula: where n i is the quantity of species i and S is the number of species.

| Macrophytes
The macrophyte samples were carefully washed with distilled water at least three times. Then, the number of leaves in each sample was quantified (including the selected leaf for area measurement and algae collection). All samples were then dried to a constant weight in a drying oven at 60°C. The dry weight of biomass of the submerged macrophytes was determined using an electronic scale.

| Snail
All snail individuals (adults and offspring) were collected from the aquariums and the quantity and fresh mass were determined. Before weighing, the snails were drained and allowed to dry on absorbent paper for 5 min (wiping the surface of snails and letting the liquid drain from their body) and then gently blotting until the surface was dry to ensure consistency among the samples (Yang et al., 2020).

| Data analyses
We used two-way ANOVA to test for the effects of macrophytes, grazing treatment, and their interaction on the environmental factors (i.e., T, DO, Cond, pH, TN, TP, NH 3 -N, and COD), followed by the least significant difference (LSD) post hoc test. The relative growth rate (RGR) of the macrophytes and snails was calculated ac- where W f (g) and W i (g) are the average final and initial mass of the snails or macrophytes in each aquarium, respectively, in grams (Gu et al.,2018). The effects of macrophyte species, grazing treatment, and their interaction on the biomass and RGR of macrophytes were determined using two-way ANOVA with post hoc LSD tests for multiple comparisons. The data describing the characteristics of snails (i.e., number, biomass, and RGR, at the total species level) from macrophytes were evaluated using one-way ANOVA with post hoc LSD tests for multiple comparisons. Two-way ANOVA was used to assess macrophyte and snail species effects on snail characteristics (i.e., number and biomass at the species level), and post hoc LSD tests were conducted for multiple comparisons. The effects of macrophytes, grazing treatment, and their interaction on phytoplankton biomass (Chl-a concentrations and Chl-a concentrations in the water were used as surrogates for phytoplankton biomass.) and epiphytic algae numeral traits (richness and abundance) were investigated using two-way ANOVA with post hoc LSD tests for multiple comparisons.
To determine the relative importance of the direct versus indirect effects of snails on macrophytes, we built a structural equa-  Table   S4C,D). The Spearman rank correlation coefficient was used to assess the correlation between four species of snails (Hellinger transformed biomass) and environmental factors.
To ensure that the data conformed to the assumptions of a normal distribution and homogeneity of variance, some parameters were log10 transformed before performing ANOVA, SEM, PCA, or RDA. Statistical analyses were performed using r version 3.6.3 with the packages agricolae (Mendiburu, 2018), vegan (Oksanen et al., 2018) and lavaan (Oberski et al., 2014), and the significance level was set to p < .05.

| Variations in water environmental factors
During the experiment, the concentrations of DO, Cond, Turb, TN, TP, NH 3 -N, and COD were notably affected by both submerged macrophyte species and snail presence (p < .05, Table 1).
The presence of snails consistently led to significantly lowered concentrations of nutrients (i.e., TN, TP, NH 3 -N, and COD) in the water associated with the three macrophyte species (p < .001,

| Macrophyte
The RGRs of the three species of macrophytes were markedly affected by their species and snail presence (Table 3, p < .05), but the interactions between these two variables were nonsignificant.
Snails significantly led to an increase in the RGR of the H. verticillata, V. natans, and E. nuttallii, (Figure 1), with H. verticillata having the greatest RGR among the three submerged macrophyte species when snails were present (Figure 1).

| Snails
The number of individuals and RGR of the snail species were markedly affected by the macrophyte species (Table 4, p < .001). The increase in number and RGR was greatest during the experiment in the presence of H. verticillata (Figure 2a,b).
The biomass and number of the four species of snails (i.e., B. aeruginosa, H. cantori, P. striatulus, and R. swinhoei) were notably affected by macrophyte and snail species identity (Table 5, p < .001).

Significant interactions between macrophytes and snail species
were observed for four snail species (  (Figure 2c,d).

| Phytoplankton and epiphytic algae
The Chl-a concentration was markedly affected by the submerged macrophyte species and snail presence (Table 5, p < .001), and there was a significant interaction between macrophyte species and the presence of snails in terms of the Chl-a concentration (Table 5 (Table S1; Figure S2). Diatoms and green algae accounted for most of the epiphytic algae ( Figure 3d). When snails were present, the abundance of diatoms and green algae tended to decrease (Figure 3d; Table 6).

| DISCUSS ION
Snails positively affected submerged macrophyte growth and development by increasing biomass, as demonstrated in both simulation experiments and field investigations (Li et al., 2008;Mormul et al., 2018;Yang et al., 2020). We found that the presence of snails significantly reduced the biomass of epiphytic algae and phytoplankton ( Figure 3). Earlier studies showed that shading by epiphytic algae and phytoplankton might limit the growth of submerged macrophytes TA B L E 2 Comparison of environmental factors associated with macrophyte and snail grazing treatments during the experiment on the basis of water temperature (T), dissolved oxygen (DO), turbidity, total nitrogen (TN), total phosphorus (TP), ammonia nitrogen (NH 3 -N), and chemical oxygen demand (COD  (Arthaud et al., 2012;Song et al., 2017;Tóth, 2013); hence, grazing by snails should favor macrophyte growth by decreasing the competition for light among epiphytic algae, phytoplankton, and submerged macrophytes (Hidding et al., 2016;Yang et al., 2020). SEM results showed that snails decreased the epiphytic algae (C = −0.38, p < .001) and phytoplankton (C = −0.69, p < .001) biomass through the improvement of macrophyte RGR (Figure 4). In addition, the pathway from the snail to macrophytes was nonsignificant (Figure 4), On the other hand, in the snail-present treatment, the nutrients in the water were significantly lower than those in the snailabsent treatment ( Table 2). The snail community might eliminate the competition between epiphytic algae and phytoplankton with macrophytes for resources (light and nutrients), and a large amount of nutrients in the water column are absorbed by macrophytes to supply their growth and reproduction (Cao et al., 2018;Kuiper et al., 2017;Li et al., 2019). Furthermore, the increase in macrophyte biomass could inhibit epiphytic algae and phytoplankton by enhancing competition for resources (light and nutrients) (Jones et al., 2000;Kuiper et al., 2017). We also found that increasing macrophyte biomass could increase the species richness of epiphytic algae (R = .43, p = .008; Figure S3), possibly by providing F I G U R E 2 Comparison of the total snail number (a), total snail relative growth rate (b), and the number (c) and biomass ( Toporowska et al., 2008).
In this experiment, both the number and biomass of the snail communities were greatest on H. verticillata (Figure 2 Figure 2). Although the leaf structure of E. nuttallii is more complex than that of V. natans, while the number and biomass of the snail communities on E. nuttallii were lower than those on V. natans in this study (Figure 2). This possibly occurred because E. nuttallii is an exotic species (Xie et al., 2010;Xiong et al., 2008). Native predators have gradually adapted to the defence strategies of native plants over long-term coevolution, while they are naive to the defence strategies of foreign plants and thus prefer to feed on native plants (Keane & Crawley, 2002;Xiong et al., 2008). Native macrophytes have a long history of coevolution with native snails, which could help snails quickly adapt to habitats containing native macrophytes.
On the other hand, the richness and abundance of epiphytic algae on H. verticillata (native) was significantly greater than that on E. nuttallii    (Figure 5a,c), namely, epiphytic algae and phytoplankton were the main food sources for B. aeruginosa (Han et al., 2010;Li et al., 2008;Zhu et al., 2013). The biomass of R. swinhoei was significantly positively correlated with the macrophytes in this study (Figure 5a,c), which indicates that R. swinhoei mainly fed on submerged macrophytes (Li et al., 2006;Li et al., 2009;Yang et al., 2020). Furthermore, we observed the R. swinhoei scraped the surface of the submerged macrophytes ( Figure S1D,E), which suggested that the R. swinhoei might graze submerged macrophytes. R. swinhoei has a large and dense radula that makes it easy to scrape and feed on the plant tissues (Xiong et al., 2008) and easily feeds on algae. Previous studies also verified that R. swinhoei feeds on macrophytes, periphytons were found to be the main food source for this species (Li et al., 2008). Previous studies verified that B. aeruginosa feeds only on algae and scrap (Li et al., 2019), mainly because its radula is small enough not to damage plant tissue. First, we hypothesized that the greater the food supply was, the greater the biomass of the snails. Second, as the correlation matrix shows, the biomass of B. aeruginosa was significantly positively correlated with epiphytic algae and phytoplankton, and the biomass of R. swinhoei was significantly positively correlated with macrophytes ( Figure 5c). According to the above, we concluded that B. aeruginosa mainly feeds on algae and that R. swinhoei mainly feeds on macrophytes. Competition has been identified as underlying niche divergence (Hardin, 1960); when predators have the same niche and multiple food sources, competition drives them to change their feeding preferences to achieve coexistence (Kolsch & Kubiak, 2011;Zaret & Rand, 1971). Consequently, competition drives snails to change their grazing preferences to achieve coexistence.

| CON CLUS ION
Snail communities can reduce the biomass of phytoplankton and epiphytic algae and thereby enhance the growth of submerged macrophytes. Macrophytes with complex architecture support more snails and epiphytic algae, and snails prefer to feed on native plants.
Competition drives snails to change their grazing preferences to achieve coexistence.

ACK N OWLED G M ENTS
This study was financially supported by the Major Science and Technology Program for Water Pollution Control and Treatment (2015ZX07503-005) and the Special Foundation of National Science and Technology Basic Research (2013FY112300). We thank Fei Ma and Jingwen Hu for their assistance during sampling. In addition, we are grateful for helpful comments and suggestions by the editor and the anonymous reviewers.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

O PE N R E S E A RCH BA D G E S
This article has earned an Open Data Badge for making publicly available the digitally-shareable data necessary to reproduce the reported results. The data is available at [https://doi.org/10.5061/ dryad.dz08k prxf].

DATA AVA I L A B I L I T Y S TAT E M E N T
All data used in the production of this article are available via Dryad: https://doi.org/10.5061/dryad.dz08k prxf.