Toxicity of a Common Glyphosate Metabolite to the Freshwater Planarian (Girardia tigrina)

Abstract

To establish meaningful policy directives for sustainable agrochemical use, we require baseline knowledge of the impacts of agrochemicals on non-target organisms. The widespread use of the herbicide glyphosate has resulted in the global presence of its metabolite, aminomethylphosphonic acid (AMPA). AMPA is commonly found in water bodies, including freshwater systems. We investigated the effects of AMPA exposure on the survivorship, regenerative abilities, and locomotion of the brown planarian (Girardia tigrina), a water-dwelling flatworm commonly found in freshwater ecosystems. In a series of experiments, we bisected and then exposed planarians to realistic field doses of AMPA for seven days and then fourteen days. For the 14-day experiment, we exposed planarians to two concentrations consistent with the high and low ranges of concentrations observed in water systems. Compared to the control group, we found that planarians exposed to AMPA for fourteen days (un-bisected for the first seven days and recovering from bisection for the subsequent seven) exhibited slower regeneration from the tail segment. Our findings highlight the potential ecological impacts of AMPA contamination on planarian populations. Quantifying the effects of AMPA exposure on planarians contributes to our understanding of the ecological consequences of our current and common agricultural practices on our freshwater ecosystems.

1. Introduction

Growers apply glyphosate-based herbicides to crops that have been genetically modified to withstand glyphosate or to unmodified crops that require timed ripening (e.g., green burndown) [1,2,3]. Once introduced into the environment, glyphosate degrades primarily via microbial action, and its major metabolite is AMPA, aminomethylphosphonic acid [4]. Glyphosate-based herbicides are currently the most frequently used pesticide in the agricultural sector [1,2,3,4]. To establish meaningful policy directives for sustainable agrochemical use, we require baseline knowledge of the impacts of agrochemicals on non-target organisms. While the toxicity of glyphosate and some of its formulations on non-target animals is well studied [5,6,7,8,9,10,11], the same is not true for AMPA; its impacts on many classes of non-target organisms are not well studied (but see [12]).

AMPA has been detected in groundwater samples in Europe, the United States, Canada, Argentina, and China [13] and in agricultural leachates of South America, Europe, and East and South Asia [14]. AMPA has been detected in various global fresh waterways, including ditches and drains, precipitation, rivers, streams, lakes, ponds, wetlands, and groundwater [4,14,15,16]. In the US, one study found AMPA in 90% of stream samples [4]. Another study found AMPA in 14.3% of sampled groundwater samples, 71.6% of sampled streams, 80.7% of sampled ditches and drains, and 29.8% of sampled lakes and ponds [4]. AMPA is water soluble and has an aquatic half-life that ranges from 2 to 91 days [17,18]. Concentrations of AMPA in freshwater environments for Mississippi and Iowa are reported at 0.02–5.7 μg/L [19]. The concentration of AMPA in Lake Balaton, Hungary, is 2.0 μg/L [20]. Median AMPA concentrations in US waterways range from 0.02 μg/L (groundwater, lakes, ponds, and wetlands) to 0.43 μg/L (ditches and drains) [4]. Median AMPA concentrations are 0.15 μg/L in waterways, with a median maximum of 5.6 μg/L [15].

Because of the ubiquity of AMPA in freshwater, it has the potential to impact non-target water-dwelling invertebrates, and this means that our current method of controlling crop weeds and harvesting crop plants can reverberate through freshwater ecosystems. Concurrently, several highly diverse taxa of aquatic invertebrates are still poorly described and are poorly considered in protection programs, despite the fact they provide a fundamental component of biodiversity [21]. We have a critical gap in need of data.

Freshwater ecosystems cover less than 1% of the planet’s surface but support up to 10% of known species. Around 25% of freshwater invertebrate species are under threat of extinction. Such a decline in species richness is likely to lead to adverse effects on the delivery of ecosystem services [22]. Planarians, for example, are small, free-living flatworms that provide ecosystem services in part by consuming small living invertebrates and decaying organisms [23,24]. They can control population growth in species such as mosquitos and snails [24,25]. They impact trophic levels as both predator and prey and contribute to sediment formation and sediment-dwelling microbial communities through bioturbation [26].

Water-dwelling planarian species are sensitive to organic matter pollution and water quality [21,23], and therefore serve as optimal bioindicators for freshwater ecosystems [27]. Additionally, planarians display an array of sensitive and reliable responses to environmental stressors, making them ideal as ecotoxicological model species [28]. Their utility in ecotoxicology studies additionally benefits from their startling developmental plasticity and regenerative capacity in response to variable nutrient conditions or injury [29,30], allowing researchers to assess the effects of potential toxins on regenerative processes.

In summary, the widespread presence of AMPA in aquatic ecosystems raises significant concerns for current agricultural methods. Our study seeks to bridge the existing gap in understanding the ecological repercussions of AMPA exposure on freshwater ecosystems. The lack of comprehensive data on the diverse taxa of aquatic invertebrates underscores the urgency for a more nuanced approach to conservation. This research addresses this gap by exploring the responses of water-dwelling planarians, shedding light on their intricate roles in maintaining ecological balance and serving as sentinel species in assessing the broader impacts of evolving agricultural practices on freshwater biodiversity. We conducted a series of experiments to determine the impact of both 7-day and 14-day AMPA exposures, using contamination levels consistent with those found in nature, on the health of a common freshwater invertebrate: the brown planarian (Girardia tigrina), a freshwater planarian.

2. Materials and Methods

As an overview, to understand the impact of current agricultural methods on freshwater invertebrates, we measured the impact of both 7-day and 14-day exposure to AMPA on the survivorship, regenerative abilities from head segments, regenerative abilities from the tail segments, regenerative abilities of the eyespots, and locomotion after regeneration in the planarian. The following information details how we executed each of the experiments.

2.1. Planarian

Brown planarians, Girardia tigrina, were purchased from Carolina Biological Supply and housed collectively in glass jars (45 × 28 × 17 cm) filled with water from our greenhouse pond. They were fed approximately 0.2 g of ground earthworm (Eisenia fetida) weekly. Following Byrne (2018) [31], we maintained them in Conviron Growth Chambers for 24 h of darkness at 23 °C. We habituated them to this housing environment for two weeks before experimentation.

2.2. Contamination

We purchased AMPA (99% pure) from Sigma-Aldrich^® (St. Louis, MO, USA) Solutions and mixed AMPA with greenhouse pond water to create two concentrations, 0.02 µg/L (low dose) and 3.1 µg/L (high dose). We selected these doses because they represent the low and high ends of the range observed in existing US water systems [4,15]. Concentrations from European waterways fall in the middle [20].

2.3. Swimming Speed

We built an apparatus to measure the speed at which individuals swam away from a bright light source. The apparatus consisted of an opaque plastic chamber (200 × 140 × 60 mm) covered with a one-way tint film. We cut a hole through the film and shone a 1000-lumen flashlight through it, creating a ring of light with a diameter of 31.65 mm. We centered a 90 mm Petri dish beneath the light and placed a sheet of graph paper beneath the dish. With the flashlight clamped 152 mm above the chamber, we set a planarian in the center of the Petri dish using a dropper and measured the time in seconds used by the planarian to leave the ring of light. Each square on the graph paper measured 2.5 mm, allowing us to measure distance per second.

2.4. Body Measurements

We used ImageJ (version 1.53g), open-source software that processes images, to assess: (1) eyespot distance (mm), defined as the distance between the center of the two pupils; and (2) body length, defined as the distance from the tip of the head to the end of the tail.

2.5. Seven-Day AMPA Exposure

As shown in Figure 1, for the 7-day exposure experiments, we created 26 replicates within two treatments: control and AMPA. Jars in the control treatment received 300.7 mL of uncontaminated pond water; jars in the second treatment received 300.7 mL of high-dose AMPA solution.

Regarding planarians, we allowed 52 group-living subjects to consume 0.2 g of shredded earthworm (Eisenia fetida) for one hour. We then extracted individuals one at a time, placed each into its own Petri dish with a dropper, and measured body length (mm). We assigned planarians to treatment groups so that the mean body lengths in both groups were similar. We then bisected each organism transversely at approximately 50% of its length, discarded the head segment, and allowed the tail segment to regenerate in its treatment for seven days, after which we assessed survivorship, swimming speed, head regeneration, and eyespot regeneration, as depicted in Figure 1.

We assessed survivorship by counting how many tail segments within each treatment survived the 7-day exposure period. To assess the proportion of body length regeneration, we divided the length of the regenerated tail segment by its length before bisection. To assess eyespot regeneration, we compared the average distance between eyespots across the two treatments. To assess swimming speed (mm^−s), we measured the speed of light avoidance for all individuals after seven days of exposure and compared means.

2.6. Fourteen-Day Exposure

As shown in Figure 2, for the 14-day exposure experiments, we created 20 replicates within three treatments: control, low-dose AMPA, and high-dose AMPA. Jars for the control treatment received 503.2 mL of pond water; jars for the low-dose experimental treatment received 503.2 mL of AMPA solution at the low-dose concentration; jars for the high-dose experimental treatment received 503.2 mL of AMPA solution at the high-dose concentration.

Before we began the experiments, we allowed 60 group-living subjects to consume 0.2 g of shredded earthworm (Eisenia fetida) for one hour. We then extracted individuals one at a time, placed each into its own Petri dish using a dropper, and measured its body length using imageJ, ensuring that mean body lengths were similar across treatments.

We did not feed the planarians for the duration of this experiment. After seven days of exposure to their treatments, we bisected them transversely at approximately 50% of their length, measured the resulting segments, and placed the resulting segments individually into jars containing uncontaminated pondwater or AMPA solutions matching their original treatment. The segments were allowed to regenerate in their appropriate solutions for an additional seven days, as shown in Figure 2.

As endpoints, we measured survivorship, body growth, degree of head regeneration, and degree of tail regeneration after 14 days of exposure. To assess survivorship, we counted the number of survivors after the first week of exposure. Additionally, we counted the number of head segments and tail segments that survived regeneration after the second week of exposure. To assess growth, we compared the initial body size to the body size measured after seven days of exposure, just before bisection. Lastly, to measure head regeneration, we subtracted the final regenerated size from the original size, and to measure tail regeneration, we subtracted the final regenerated tail size from the original.

2.7. Data Analysis

All data were recorded in Microsoft Excel. We ran all analyses in StatPlus v6. Samples size (N), means, and standard deviations are reported throughout. Error bars in graphs represent the standard error of the means. After determining if variables were distributed normally using a Kolmogorov–Smirnov/Lilliefor test, we examined differences between means using one-way ANOVAs significant at the 0.05 level. When we detected a difference in means, we followed up with post hoc Scheffe tests, which are designed for unplanned comparisons, to determine where the differences lay. For data that were not normally distributed, we used a non-parametric Kruskal–Wallis ANOVA significant at the 0.05 level. To assess the impact of contamination on survivorship, we used a chi-squared test.

3. Results

3.1. Ensuring Equality before the Inception of the 7-Day Exposure Experiment

Individuals slated for the uncontaminated water (N = 26) had a mean and standard deviation in body length of 10.61 ± 1.94 mm, and individuals slated for the AMPA-contaminated water had a mean and standard deviation in body length of 10.71 ± 1.93 mm. A one-way ANOVA detected no significant difference (DF = 1, 50; F = 0.04; p = 0.84).

3.2. Survivorship after 7-Day Exposure

During the 7-day-exposure event, regenerating in contaminated water did not impact survivorship. Tail segments that regenerated in uncontaminated pond water demonstrated 92.3% (24/26) survivorship, while those that regenerated in AMPA-contaminated water demonstrated 96.2% (25/26) survivorship. These differences were not significant according to a chi-squared test (X² = 0.08; DF = 1; p = 0.78).

3.3. Body Regeneration after 7-Day Exposure

Regeneration was not slowed by a 7-day exposure to AMPA during regeneration. After bisection and exposure, tail segments in the control group (N = 24) regained a mean of 65.0 ± 9.4% of their initial body size, and those in the AMPA treatment (N = 25) regained 65.7 ± 10.2% of their initial size. An ANOVA found no difference (DF = 1, 47; F = 0.07; p = 0.79).

3.4. Eye Spot Regeneration after 7-Day Exposure

Seven days of exposure to AMPA during regeneration did not alter regenerated head morphology, as measured by the distance between eyespots. Before bisection and exposure to contamination, the mean distance and standard deviation between eyespots was 0.56 ± 0.12 mm for the control group (N = 26) and 0.56 ± 0.11 mm for the group exposed to AMPA (N = 26). After bisection and seven days of exposure, planarians in uncontaminated water (N = 24) regained 56.1 ± 16.6% of their initial value, while those regenerating in AMPA-contaminated water regained 56.5 ± 13.9% of their initial value. An ANOVA found no significant difference (DF = 1, 47; F = 0.007; p = 0.93).

3.5. Swimming Speed after 7-Day Exposure

Seven days of exposure to AMPA during regeneration did not alter locomotory ability, as measured by swimming speed when exposed to light. Planarians that were bisected and then returned to uncontaminated water for seven days (N = 24) demonstrated a mean swimming speed of 0.75 ± 0.22 mm^−s. Similarly, planarians that regenerated in AMPA-contaminated water (N = 25) demonstrated a swimming speed of 0.83 ± 0.22 mm^−s. An ANOVA detected no significant difference (DF = 1, 48; F = 1.61; p = 0.21).

3.6. Ensuring Equality before Inception of the 14-Day Exposure Experiment

Before we began this experiment, we ensured that body lengths were equal across treatments. We found the following distribution of body lengths: individuals in the control treatment (N = 20) measured 9.40 ± 2.39 mm; individuals in the low-dose treatment (N = 20) measured 10.36 ± 4.79 mm, and individuals in the high-dose treatment (N = 20) measured 8.43 ± 1.62 mm. A non-parametric Kruskal–Wallis ANOVA detected no significant difference in initial mean body lengths across treatments (DF = 2; H = 3.29; p = 0.19).

After the initial exposure period, we bisected each planarian before re-exposing it for another week. Via this bisection process, we produced heads with the following means and standard deviation: a total of 3.85 ± 1.24 mm for the control (N = 20); 3.47 ± 1.13 mm for the low-dose treatment (N = 20); and 3.70 ± 1.24 mm for the high-dose treatment (N = 20). These lengths did not significantly differ from each other (DF = 2, 57; F = 0.51; p = 0.60). Via bisection, we produced tails with the following means and standard deviations: a total of 2.80 ± 1.14 mm for the control (N = 20); 2.70 ± 0.70 mm for the low-dose treatment (N = 20); and 3.17 ± 1.35 mm for the high-dose treatment (N = 20). According to a Kruskal–Wallis ANOVA, these lengths were not significantly different from each other (DF = 2; H = 1.43; p = 0.49).

3.7. Survivorship While Regenerating during 14-Day Exposure

For this experiment, we measured the survivorship of planarians at two time points: (1) after the initial 7-day exposure period but before bisection, and then (2) after bisecting them and exposing them for seven additional days. All planarians survived the initial 7-day exposure period, demonstrating 100% survivorship (20/20), regardless of treatment.

After we bisected those planarians and let them regenerate for one week, mortality occurred only in the treatments with contamination. For the planarians regenerating in uncontaminated water, survivorship of the head segments was 100% (20/20). Survivorship of the head segments regenerating in the low-dose AMPA treatment demonstrated a 95% (19/20) survivorship, and the head segments regenerating in the high-dose AMPA treatment demonstrated a 90% (18/20) survivorship. These differences were not significant according to a chi-squared test (X² = 0.11, DF = 2, p = 0.95).

Similarly, the tail segments regenerating in the uncontaminated pond water demonstrated a 100% (20/20) survivorship; those regenerating in the low-dose AMPA treatment demonstrated 85% (17/20) survivorship; and for the planarians regenerating in the high-dose AMAPA treatment, the tail segments demonstrated 85% (17/20) survivorship. These differences were not significant according to a chi-square test (X² = 0.33, DF = 2, p = 0.85).

3.8. Growth during 14-Day Exposure

This experiment began with seven days of exposure before bisection and seven additional days of exposure after bisection. After the initial exposure period, all planarians demonstrated reduced body length, likely because they received no food. Planarians living in uncontaminated water (N = 20) decreased by 1.32 ± 2.07 mm in length. Planarians living in low-dose AMPA (N = 20) decreased by 3.16 ± 5.03 mm, and planarians exposed to high-dose AMPA (N = 20) decreased by 1.64 ± 1.49 mm. While planarians living in uncontaminated water decreased in body length the least, a non-parametric Kruskal–Wallis ANOVA detected no significant difference (DF = 2; H = 1.30; p = 0.52).

3.9. Head Regeneration after 14-Day Exposure

After bisection and seven more days of exposure, head-segment lengths increased in all treatments. The head segments in the control group increased the most, demonstrating a mean and standard deviation increase of 1.34 ± 1.08 mm (N = 20). The head segments regenerating in the low-dose AMPA water increased 1.01 ± 1.08 mm (N =19), and the head segments regenerating in high-dose AMPA water increased 0.83 ± 1.31 mm (N = 18). While the control group had the greatest recovery, an ANOVA detected no significant difference in head-segment regeneration (DF = 2, 54; F = 0.96; p = 0.39).

3.10. Tail Regeneration after 14-Day Exposure

Fourteen days of AMPA exposure significantly slowed tail segment regeneration (ANOVA: DF = 2, 51; F = 4.89; p = 0.01), particularly at the high-dose level. Tail segment length increased in all treatments, with the control group demonstrating a mean and standard deviation increase of 1.55 ± 1.27 mm (N = 20). Individual tail segments within the low-dose treatment demonstrated a mean body length increase of 1.27 ± 1.38 mm (N = 17), and tail segments regenerating in the high-dose AMPA water increased a mean of 0.29 ± 1.14 mm (N = 17). A Sheffe post hoc test identified a significant difference (p = 0.02) between the means for the control and the high-dose treatment; the difference between the means for low and high approach significance (p = 0.08). See Figure 3.

4. Discussion

This series of experiments showed that 7-day AMPA exposure had no impact on our measured endpoints, and 14-day exposure had no impact on regeneration from head segments or general body growth; however, 14 days of exposure to high but realistic doses of AMPA significantly slowed regeneration from tail segments and had a negative impact on planarian mortality. Taken together, these results suggest planarians can likely recover from one short exposure but that multiple exposures or chronic exposure to AMPA will impact the ecosystem services delivered by planarians.

Given the ubiquity of AMPA in freshwater systems resulting from current agricultural practices [4], our results suggest the need to assess the response of planarians living in the wild to actual chronic field doses of AMPA. Several taxa of aquatic invertebrates, including planarians, are still poorly described [21]. They are rarely considered in protection programs [21]. In our laboratory experiments, the 14-day exposure negatively impacted the planarians’ ability to regenerate. In the wild, exposure lengths are likely to exceed two weeks, and exposure is likely to co-occur with glyphosate [4]. Planarians provide a fundamental component of biodiversity [21] and deliver critical ecosystem services. Understanding the impact of this contaminant on these invertebrates is vital.

Furthermore, our experiments used conservative doses of AMPA, with our high-dose experiments not quite reaching the highest observed AMPA levels in lakes, streams, and ponds, and our longest exposure period lasting only two weeks. When planarians are exposed to AMPA for time periods extending beyond two weeks, they might endure more dramatic health impacts. Follow-up experiments might include water contaminated with both AMPA and glyphosate.

Data about AMPA toxicity in aquatic systems are scarce and contradictory [32,33]. A literature review concerning the impact of AMPA on aquatic organisms reported serious impacts on the European eel (Anguilla anguilla), zebrafish (Danio rerio), guppies (Poecilia reticulata), Mediterranean mussels (Mytilus galloprovincialis), and toads (Bufo spinosus) [12] but another study found that AMPA was the least toxic of the chemicals associated with glyphosate-based formulations to zebrafish [33]. AMPA exposure causes genotoxicity and immunotoxicity in fish, adverse changes in hemolymph parameters, effects on mussels’ antioxidant enzymes, as well as developmental delays and affecting the survival of tadpoles [12]. Also, AMPA can induce sublethal responses in mosquito (Aedes aegypti) larvae during acute exposures [32]. However, under environmentally realistic concentrations, AMPA was not toxic to aquatic vertebrates and invertebrates [34,35]. The no observed-adverse effect concentration (NOAEC) of AMPA to fathead minnow fish (Pimephales promelas) and Daphnia magna was far below levels encountered in nature [36]. Similarly, AMPA had no effect on the mortality of the earthworm (Eisenia andrei) under field-relevant concentrations [37]. In the few studies that examined the impact of AMPA on planarians, researchers found that exposure can cause seizures in D. dorotocephala via the inhibition of glutamate-activated ion channels [38,39] but both studies were aimed at understanding pharmacology rather than ecotoxicity. Our experiments add to our understanding of the impact of field-realistic doses of AMPA on an important freshwater invertebrate.

On a different note, our study finds that after the planarians lived in AMPA-contaminated water for a week and then regenerated in it, only the tail segments demonstrated a significantly slower rate relative to those regenerating in the uncontaminated water; head segments were not similarly impacted. This finding suggests that exposure to AMPA might impact axial patterning, the process by which the anterior–posterior (head–tail) axis is established and maintained during regeneration.

After bisection, position control genes (PCGs) govern the planarian patterning pathways that influence whether to grow a head or a tail [40]. Upon bisection, the planarian tail segment will express anterior PCGs at the anterior-facing wound, and the planarian head segment will express posterior PCGs at the posterior-facing wound [41]. Regional expression of Wingless/Integrated (Wnt) ligands (posteriorly) and Wnt inhibitors (anteriorly) are necessary for planarians to develop head and tail segments [42,43]. Once a wound has been inflicted, PCGs will express Wnt1, which leads to the expression of ß-catenin [42]. ß-catenin expression triggers planarians to regenerate tail segments, while Wnt inhibitor notum expression inhibits ß-catenin expression, leading to head regeneration [44].

Our findings suggest that AMPA may interfere with the expression of Wnt inhibitors, which may lead to a slower rate of head regeneration from tail segments. Future studies can employ assays to determine the expression of PCGs to further understand the impact of AMPA on axial patterning and regeneration rate.

In summary, our experiments demonstrate that exposing freshwater planarians to realistic field doses of AMPA for 14 days caused sublethal impacts on their health. Impactful follow-up studies might include using a series of AMPA concentrations to calculate the EC₅₀. Given the role that planarians play in nutrient cycling, control of other invertebrate populations, providing prey for organisms higher on the trophic levels, and in bioturbation, our findings suggest that our current agricultural reliance on glyphosate to control weeds and harvest crops may have a measurable impact on freshwater ecosystems. As we strive towards sustainable growing methods, this cost belongs in the accounting.