Multigenerational

Effects of Weathered Polyethylene Microplastics on Drosophila melanogaster

Authors: Shawninder Chahal, Jun-Ray Macairan, Laura M. Hernandez, Hoai-Nam N. Bui, Anthony Smith, Hans C. E. Larsson, Nathalie Tufenkji

Abstract

Most plastics are released into the environment once they are discarded, resulting in microplastics (<5 mm) being found in every part of the world. To better understand their toxicity, we exposed four generations of the fruit fly, Drosophila melanogaster, to high-density polyethylene (HDPE) microplastics (<38 μm) and monitored their reproductive performance. We found that the eclosion fraction of pupae was 5.7% higher (p < 0.05) in the stream of flies that were fed 100 mg/kg HDPE across all generations when compared to those fed control food. Several fourth-generation treatment flies expressed malformed salivary glands and scutella. No toxicity was observed in the number of pupae and flies that emerged from the eggs laid in each generation or their mean pupation and eclosion time. Moreover, no toxicity was observed in the development of larvae into adult flies (0.1–10 000 mg/kg HDPE) and various sublethal end points (100 mg/kg HDPE) such as larva and adult fly locomotion and the mass of female and male flies. These results, obtained with realistic weathered microplastics, indicate that HDPE microplastics at the concentrations evaluated are unlikely to be lethal in Drosophila melanogaster; however malformations are still possible despite little-to-no observable internalization of the microplastics.

Introduction

Upon disposal, it is estimated that 79% of plastic is released into the environment, including landfills. Over time, these plastics experience photo-oxidation from solar UV radiation; mechanical weathering through abrasions with sand, soil, and rocks; and to a lesser extent, thermal degradation. This natural weathering of plastics produces microplastics, which are loosely defined as plastic particles ranging from ∼1 to 5000 μm in size. Due to the widespread use of plastics, microplastics are consequently found in every corner of the planet and their concentration in the environment can vary considerably. For instance, soil microplastic concentrations in industrial areas can range from 300 to 67 500 mg/kg. Conversely, a study conducted in the Tibetan Plateau found only 48 microplastic particles per kg of soil, noting that 38% were smaller than 50 μm and that polyethylene was the most abundant, comprising 44–49% of the microplastics found in the soils.

The toxicity of microplastics in aquatic organisms has been well documented. Some of the most commonly examined microplastics in aquatic ecotoxicology studies are those derived from polystyrene, with toxicity being reported in Scrobicularia plana, Gammarus pulex, Dolioletta gegenbauri, and goldfish. However, polyethylene makes up 29% of plastic production and 32% of plastic waste, making it one of the most widely used plastic polymers. In particular, high-density polyethylene (HDPE) comprises 13% of total plastic waste. Polyethylene has also shown toxicity in aquatic organisms such as Crassostrea gigas and zebrafish. While there are many studies examining microplastic toxicity toward a diverse set of aquatic organisms, far fewer studies exist that examine the toxicity of microplastics to terrestrial organisms. This is despite the existence of studies that have demonstrated the ability of some terrestrial organisms to consume polyethylene as a form of biodegradation. For instance, Yang et al. found that Plodia interpunctella larvae can degrade polyethylene upon consumption via bacteria found in their gut. A similar phenomenon was reported in Galleria mellonella larvae, whereby it was shown that their saliva contains enzymes capable of breaking down polyethylene. These studies show that polyethylene has the potential to be more bioactive than was once assumed, especially in the larvae of insects. Therefore, it is critical to study the toxicity of this type of plastic in these organisms.

Drosophila melanogaster, commonly known as the fruit fly, has long been used as a model organism for genetics studies but is now seeing increasing use in toxicity studies. Contributing to this rising popularity is their rapid development time, typically less than 2 weeks, and relatively low maintenance, in terms of both time and cost, allowing for studies to be conducted with a large sample size and over multiple generations. They are also well-suited for reproductivity assays, whereby a female fly can produce hundreds of offspring. Moreover, approximately 75% of the human genes associated with various diseases have a similar counterpart in flies. Flies also undergo distinct development cycles starting as an egg and developing into larvae, pupae, and finally adult flies. This can allow one to isolate which stage of development an examined substance may cause toxicity in, whether toxicity is only observed after long-term exposure, or even over several generations. The study of toxicity over multiple generations of a species can reveal long-term effects that could otherwise go unnoticed in a single generation. For instance, in a study by Panacek et al. where fruit flies were exposed to 5 mg/L silver nanoparticles, no difference in the number of hatched flies was observed in the first generation, but there was a sharp decline in the second generation. Zhao et al. exposed Eisenia fetida to 1–5 w/w% low-density polyethylene microplastics over two generations, and found declines in superoxide dismutase, catalase, peroxidase, and acetylcholinesterase activity in the offspring relative to the control. However, when Chironomus tepperi was exposed to 1000 polyethylene microplastic particles per kg sediment in the parental generation, but not the offspring generation, no changes in the lipidome of the offspring were observed relative to the control.

Sublethal toxic effects have been observed in Drosophila melanogaster after microplastic and nanoplastic exposure. Matthews et al. exposed fruit flies to 20 nm and 1 μm polystyrene particles, finding that 100 ppm exposure of either size caused gastrointestinal tract damage. Drosophila melanogaster larvae exposed to 10 mg/mL polyethylene microplastics had 27% slower mean crawling speed than control larvae, and 10 percentage point fewer adult flies could climb 3 cm in 10 s than the control flies. These findings emphasize the importance of looking beyond just lethality when determining the toxicity of microplastics.

Herein, we present our ecotoxicology study on the effects of HDPE microplastics on the fruit fly, Drosophila melanogaster. We produced HDPE microplastics from bulk HDPE through accelerated mechanical and ultraviolet weathering to simulate real-world weathering of plastics. Larvae were raised on food containing HDPE microplastics at concentrations of 0.1, 1, 10, 100, 1000, and 10 000 mg/kg, covering the wide range of microplastic concentrations found in the environment, from remote locations, industrial zones, or pollution hotspots. Further experiments were conducted at 100 mg/kg food to measure the locomotion of larvae and adult flies, the mass of female and male flies, and the reproductive performance of the flies over 4 generations. Light sheet fluorescence microscopy was also performed to assess the potential uptake of the HDPE microplastics in the flies. Microcomputed tomography was performed to examine any malformations in the flies. The comprehensive nature of this work, namely, examining developmental toxicity over the span of 6 concentrations, as well as the multigenerational assessment of HDPE microplastic toxicity, provides new insights into the long-term effects of HDPE microplastics exposure and provides knowledge that can apply to various regions around the world.

Experimental Section

Preparation of Weathered High-Density Polyethylene

Microplastics

HDPE sheets (48″ × 96″ × 1/16″, McMaster-Carr 8619K112) were cut into squares of approximately 4 cm × 4 cm. The HDPE squares were first rinsed with 70 vol % ethanol to remove any oil or dust that may have accumulated on the surface through handling and storage. The HDPE squares were then rinsed with reverse osmosis water (Type II, >1 MΩ), hereafter referred to simply as water. Any residual water was left to be absorbed by a brown paper towel. The dry HDPE squares were then placed in a 2500 W, 36 000 rpm, stainless steel spice grinder (Homend) and left to run for 2 min and then turned off to allow the grinder to cool. This process was repeated by continuing to add HDPE squares and running the grinder for 2 min at a time until the grinder was approximately half full and there was no longer any visible difference in the fineness of the HDPE between runs. The granular HDPE was then passed through a 38 μm opening (US Standard Mesh No. 400) stainless steel sieve (Cole-Parmer RK-59984-24) to obtain HDPE microplastics. The HDPE microplastics (<38 μm) were then placed on an acetone-washed glass Petri dish and placed under UV irradiation (365 nm, 25 W/m^2^) for 14 days at 25 °C in an incubator (INFORS HT Multitron Pro) to induce photo-oxidation. The glass Petri dish was shaken every 2 days to ensure that the HDPE was uniformly exposed to the UV light over the 14 days. Details on HDPE characterization methods including scanning electron microscopy (SEM) and Fourier-transform infrared (FTIR) spectroscopy can be found in the Supporting Information (SI).

Maintenance of Drosophila melanogaster

Maintenance of Drosophila melanogaster (Oregon-R strain) was performed as reported previously. Flies were raised on food consisting of 84.6 wt % water, 14.9 wt % Nutri-Fly Bloomington Formulation powder (Diamed, GEN66-112), and 0.5 wt % sodium propionate (Genesee Scientific, 20-271) in glass culture bottles (Carolina, 173135) in an environmental chamber (Panasonic MLR-352H-PA) operating under a day/night cycle (12 h/12 h) at 60% relative humidity at 25 °C. Illumination during the day cycle was ∼1500 lx and night cycles were complete darkness.

To ensure that the flies for any given experiment were of similar age, egg collection cages were set up. Approximately 5 g of yeast paste (2.5 g dry yeast (Diamed, GEN62-106) mixed with 2.5 g water) was spread on the center of a grape agar (Diamed, GEN47-102) plate. A fly cage (Diamed, GEN59-101) was placed on top of the agar plate where the flies laid eggs and marked the starting point of most experiments (i.e., day 0). After 4 h, the flies were removed from the cage and the eggs were allowed to age for an additional 24 h. Any larvae that emerged from the eggs in the agar plate were then used for subsequent experiments unless stated otherwise.

When a particular selection of flies was necessary (e.g., sorting of female and male flies), flies were anesthetized with CO2 on a flypad (Genesee Scientific, 59-114) and separated as needed. To minimize any detrimental physiological effects on the flies, the total anesthesia time was limited to less than 10 min.

Developmental Toxicity

Developmental toxicity was measured according to previous methods with some modifications. , First, 10 mL of food containing 0, 0.1, 1, 10, or 100 mg HDPE microplastics per kg total food was dispensed into a 50 mL glass test tube. The food preparation required a step where the food was first brought to a boil, simmered for 10 min, then cooled down. The HDPE microplastics were mixed into the food after the simmer phase as the temperature began to decrease to minimize any additional weathering that could have occurred at such an elevated temperature. Once the food had cooled to room temperature and had solidified, approximately 20 first instar larvae were transferred from the grape agar plates to the test tube which was then sealed with a cotton plug. For each test tube, the number of pupae and flies were counted every 2 days over 14 days. If any flies were present, they were removed from the vial, and not returned, to facilitate counting. Each treatment was done in up to four replicates (i.e., four test tubes). The entire experiment was then repeated three times for a total of up to 12 replicates (i.e., N ≤ 12). This assay was performed again with food containing 0, 1000, or 10 000 mg/kg HDPE. In this case, each treatment was done in six replicates (i.e., six test tubes). The entire experiment was then repeated twice for a total of 12 replicates (i.e., N = 12).

Multigenerational Reproductive Toxicity Assay

Multigenerational toxicity was measured according to previous methods with some modifications. First, 10 mL of food containing 0 or 100 mg HDPE microplastics per kg total food was dispensed into a 50 mL glass test tube. Once the food had cooled to room temperature and had solidified, approximately 20 first instar larvae were then transferred from the grape agar plates to the test tube and the test tube was then sealed with a cotton plug. On day 9, any flies present in the test tube were released from the test tube. After 4 h, the newly emerged flies (F0, parental flies) were separated into mating pairs (one female, one male) and placed into a new test tube containing 10 mL of food with either 0 or 100 mg HDPE microplastics per kg total food.

This step reset the clock to day 0 for the new test tube since the F0 flies had entered a new test tube in which they would begin to lay eggs. On day 4, the F0 mating pair was removed from the test tube. The eggs laid by the F0 flies gave rise to the first generation (F1) of flies. On day 12, any F1 flies present in the test tube were released from the test tube. After 4 h, the newly emerged F1 flies were separated into mating pairs (one female, one male) and placed into a new test tube containing 10 mL of food with either 0 or 100 mg HDPE microplastics per kg total food.

The process described in the preceding paragraph was continued for the second (F2), third (F3), and fourth (F4) generation of flies. For each test tube, the number of pupae and flies were counted every 2 days for 16 days. If any flies were present, they were removed from the vial, and not returned, to facilitate counting. Each treatment was done in up to six replicates (i.e., six test tubes). The entire experiment was then repeated twice for a total of up to 12 replicates (i.e., N ≤ 12). When transferring flies to the next generation, up to three mating pairs may have been taken from a single test tube of the previous generation. This was done to ensure that the sample size did not decrease throughout the experiment since sometimes no potential mating pair would emerge during the 4 h eclosion period from some of the test tubes or in some cases the parent flies did not produce any offspring. Mating pairs were selected by first separating the male and female flies into two groups on the CO2 flypad under a microscope, and then randomly selecting one male and one female fly to place in the test tube without using a microscope. This was done to ensure that there was no bias toward selecting flies based on their physical appearance (e.g., potential wing damage, irregular phenotypic characteristics, etc.). Table shows an overview of the food which was fed to each generation in each treatment stream.

Larval Crawling Assay

Larval crawling was measured according to previous methods with some modifications. , First, 10 mL of food containing 0 or 100 mg HDPE microplastics per kg total food was dispensed into a 50 mL glass test tube. Once the food had cooled to room temperature and had solidified, approximately 20 first instar larvae were then transferred from the grape agar plates to the test tube which was then sealed with a cotton plug. On day 4, a larva was randomly selected from the test tube and transferred onto the center of a 100 mm grape agar Petri dish which did not contain HDPE microplastics. The larva was given 30 s to adjust to its environment and was then video recorded (Canon EOS Rebel SL2, 1920 × 1080 resolution, 29.97 frames per s) under an Olympus SZX16 stereo microscope for 60 s. The plate was slowly moved as the larva crawled such that the larva remained near the center view of the camera. The number of peristaltic contractions performed by the larva in 60 s was recorded. This was repeated with a total of 3 larvae from each test tube. The average number of peristaltic contractions per minute from these three larvae was taken as one replicate (i.e., N = 1). Each treatment was done in six replicates (i.e., six test tubes). The entire experiment was performed twice for a total of 12 replicates (i.e., N = 12).

Climbing Assay

A climbing assay was performed according to previous methods with some modifications. ,, First, 10 mL of food containing 0 or 100 mg HDPE microplastics per kg total food was dispensed into a 50 mL glass test tube. Once the food had cooled to room temperature and had solidified, approximately 20 first instar larvae were then transferred from the grape agar plates to the test tube which was then sealed with a cotton plug. On day 11, the flies that emerged from the larvae were transferred to a 100 mL glass graduated cylinder marked at a height of 10 cm which was capped with a cotton plug. The cylinder was tapped to move the flies to the bottom and the flies were then allowed to ascend. After 10 s, the number of unique flies that crossed the 10 cm mark was recorded. This cycle was repeated five times for each test tube of flies to obtain an average value recorded as the number of flies passing the 10 cm mark divided by the total number of flies and represented one replicate (i.e., N = 1). This process was video recorded (Canon EOS Rebel SL2, 1920 × 1080 resolution, 29.97 frames per s) so that counting could be done later. Each treatment was done in six replicates (i.e., six test tubes). The entire experiment was performed three times for a total of 18 replicates (i.e., N = 18).

Fly Mass Measurements

Fly mass measurements were performed as reported previously. After the flies went through the climbing assay, they were anesthetized using CO2, separated into female and male groups, and the collective mass of each sex was divided by the number of flies for each sex from each test tube and represented one replicate (i.e., N = 1). Each treatment was done in up to six replicates (i.e., six test tubes). The entire experiment was performed twice for a total of up to 12 replicates (i.e., N ≤ 12).

Fly Imaging

Light sheet fluorescence microscopy was performed as reported previously, ,, with some modifications. Micro-computed tomography (μCT) was also performed according to previous methods, with some modifications. Tomographic μCT images were qualitatively analyzed, and 3D models were reconstructed with Dragonfly image analysis software (Object Research Systems Inc.). Detailed methodologies for light sheet fluorescence microscopy and micro-computed tomography can be found in the SI.

Statistical Analysis

All data analysis was performed in Python 3.9.12 using numpy 1.21.5, pandas 1.4.2, , scipy 1.7.3, and statsmodels 0.13.2. Plots were made using matplotlib 3.5.1. , In this work, p < 0.05 was considered statistically significant. Detailed methodologies on the statistical analysis can be found in the SI.

Results and Discussion

Characterization of HDPE Microplastics

The HDPE microplastics were imaged using SEM and dark field microscopy. A 38 μm mesh sieve was used to isolate the smaller microplastics which were used in the exposure studies. The use of a sieve implies that only two dimensions of any particle or aggregate had to be below this size allowing for larger particles to pass through in a third dimension. Figure a shows that when the microplastics were well dispersed, many nanosized plastic fragments were present. These nanosized plastics had a median size of 228 nm (Figure a, inset). However, many larger microplastics were present as well. Figure b reveals that these larger particles may often be aggregates of many smaller microplastics. Figure c shows a visible-light image of the HDPE microplastics under dark-field microscopy (Olympus BX43). Using visible light allows one to see the translucence of the microplastics, facilitating the visualization and thickness of these particles as they twist and bend.

These images, taken under various microscopy methods and magnifications, indicate that there is a wide distribution of sizes found in our microplastics. On a mass basis, it is likely that most of the microplastics have a size closer to ∼38 μm, whereas on a number basis most may be ∼200 nm. Moreover, although aggregation was observed under SEM, the dispersibility of the microplastics may be considerably different than during the food preparation whereby microplastics were added as the food mixture was cooling down from a boil. The food mixture, despite being above room temperature at the time of HDPE addition and consisting mostly of water, had a high viscosity and it is unclear to what extent this may affect dispersion. Finally, upon consumption of microplastics by Drosophila melanogaster, the microplastics may further aggregate or disperse based on the unique chemical profile in the fluids contained in various portions of the gut.

To assess the chemical composition of the plastic samples, FTIR spectroscopy was performed. Three prominent peaks that are characteristic of HDPE were observed (Figure S1). Peaks associated with C–H stretching of −CH2– groups were found at 2850 and 2912 cm^–1^, while a peak at 1470 cm^–1^ originated from C–H bending. We note that upon UV weathering, there were not any considerable changes to the microplastics with regards to the chemical functional groups observed. The FTIR spectra of the filtered HDPE suspension confirmed that any sub-200 nm particles found in the HDPE suspension were, despite having a weaker signal, showing the same distinctive peaks as HDPE.

Developmental Toxicity

A major factor contributing to the popularity of Drosophila melanogaster as a model organism is its rapid development time from egg to adult. To measure the toxicity of HDPE microplastics on the development of Drosophila melanogaster, batches of approximately 20 first instar larvae, were transferred to test tubes containing HDPE microplastics in concentrations of 0.1, 1, 10, and 100 mg/kg food, and in a separate set of experiments, HDPE concentrations of 1000 and 10 000 mg/kg food. Both sets of experiments had their respective controls and were conducted separately from each other.

Figure a,b shows that there were no significant differences in the total number of pupae or flies that emerged from the larvae in the HDPE concentration range of 0–100 mg/kg. A two-way ANOVA revealed that no significant differences based on concentration were detected in the total pupae (p = 0.174) or flies (p = 0.133), even after accounting for significant differences based on experimental block in the total pupae (p = 0.011) and flies (p = 0.006). Similarly, at 0, 1000, and 10 000 mg/kg HDPE in food, Figure a,b shows no significant differences on the basis of concentration (total p = 0.079; total p = 0.606) or experimental block (total p = 0.267; total p = 0.219). This shows that at concentrations as high as 1 wt % HDPE microplastic in food, no statistically significant toxicity was observed even with a sample size of N = 12.

The eclosion fraction measures the fraction of pupae that successfully eclosed into adult flies. An eclosion fraction considerably lower than 1 could indicate that the flies had difficulty developing into adult flies, resulting in death at the pupal stage. Given that there were no significant differences observed in the total number of pupae and flies, it followed that there were also no significant differences in the eclosion fraction (Figure S2) based on concentration or experimental block in the 0–100 mg/kg HDPE set (concentration: p = 0.249; p = 0.596) and the 0, 1000, and 10 000 mg/kg set (concentration: p = 0.538; p = 0.278). In general, the eclosion fractions were high at all concentrations with mean values in the range of 0.97–1.00, suggesting that the eclosion of Drosophila melanogaster was unaffected by HDPE microplastics at concentrations as high as 10 000 mg/kg.

The data in the developmental toxicity assay was collected every 2 days over 14 days, allowing for the calculation of mean emergence times. Figure c,d shows that there were no significant differences in the mean pupation or eclosion time of larvae in the HDPE concentration range of 0–100 mg/kg on the basis of concentration (pupation: p = 0.783; p = 0.070) or experimental block (pupation: p = 0.091; p = 0.389). Similarly, at 0, 1000, and 10 000 mg/kg HDPE in food, Figure c,d shows no significant differences on the basis of concentration (pupation: p = 0.687; p = 0.077) or experimental block (pupation: p = 0.584; p = 0.462).

The developmental toxicity of microplastics over a variety of plastic types, sizes, and concentrations has previously been studied. Drosophila melanogaster raised from egg to adult fly in 4, 10, or 20 μm polystyrene microplastics in concentrations ranging from 2.604 to 2604 mg/kg food resulted in the viability of the eggs decreasing as concentration increased with all sizes. At the highest concentration, the viability of the eggs dropped from 100% in the control to 76%, 82%, and 87%, when exposed to 4, 10, and 20 μm polystyrene microplastics, respectively, however, the study concluded that this was nontoxic. A study by Liu et al. found that exposing Drosophila melanogaster to 0.1 μm amino-functionalized polystyrene particles at a concentration of 50 μg/L resulted in no significant differences in the fraction of embryos that successfully developed into pupae or flies compared to the control. Likewise, the developmental toxicity assay performed in the present study showed that the HDPE microplastics, at concentrations as high as 10 000 mg/kg, had no measurable impact on the mortality or development time of larvae as they transform into pupae and then adults during the first 14 days of their life. The larvae of other insects have been shown to break down polyethylene. For instance, the gut bacteria of Plodia interpunctella larvae and the saliva of Galleria mellonella larvae can degrade polyethylene. Although these two species and Drosophila melanogaster are all insects with somewhat similar development cycles, P. interpunctella and G. mellonella are more closely related to each other by belonging to the same family, Pyralidae, and are notably much larger than D. melanogaster. Therefore, the lack of observed developmental toxicity in D. melanogaster could suggest that they may be unable to break down polyethylene and potentially ignore the microplastics altogether or have them pass through as inert materials during the larval phase. The lack of chromophores in polyethylene makes it more resistant to photodegradation, however small amounts of peroxides may still be produced by this mechanism. Conversely, it has been well-established that other microplastics, such as those derived from polystyrene, cause oxidative stress in organisms. ,

Multigenerational Reproductive Toxicity Assay

The developmental toxicity assay looked at toxicity over the first 14 days after the eggs were laid and found no signs of mortality or developmental delays. Multigenerational assays have the potential to detect toxicity at concentrations that would otherwise be too low for acute toxicity to be observed. For instance, Panacek et al. conducted a study evaluating the toxicity of silver nanoparticles on Drosophila melanogaster, finding that 10 mg/L silver nanoparticle exposure did not induce acute developmental toxicity. Their multigenerational study using 5 mg/L silver nanoparticles similarly resulted in no toxicity in the first generation. However, a sharp decline in the number of flies eclosed in generations 2–4, recovery in generations 5–6, and a return to levels seen in the control by generation 7, albeit with decreased fly mass, followed. To measure the long-term toxicity of HDPE microplastics on the reproductive performance of Drosophila melanogaster, mating pairs, consisting of one female and one male fly that had eclosed less than 4 h earlier, were transferred to test tubes containing control food or 100 mg/kg HDPE in food. The offspring of this mating pair were then put through the same process. This cycle was repeated for 4 generations. Three streams of food supply were set up for this assay (Table ). Stream A consisted of control food throughout all 4 generations, while Stream B consisted of HDPE food throughout. Stream C featured the initial parent flies that were used to start the experiment raised on HDPE, while subsequent generations were fed control food. The study was arranged as such to determine if the HDPE microplastics can impact the parent flies in such a way that future generations would also be impacted even when HDPE was no longer present in the food.

The developmental toxicity assay showed that 100 mg/kg HDPE in food did not impact the development of Drosophila melanogaster. Therefore, any differences in the number of pupae or flies that emerged from the mating pairs would likely be due to differences in reproductive performance between the mating pairs from different streams or another unmeasured biological factor. From each generation, the total number of pupae and adult flies that would emerge from the eggs laid by their parents after 16 days was measured. Using a three-way ANOVA with stream, generation, and experimental block as factors, Figure a,b shows that there were no significant differences in the total number of pupae (stream: p = 0.712; p = 0.339; p = 0.129) or flies (stream: p = 0.493; generation: p = 0.233; p = 0.214) that emerged throughout the assay. Interestingly, significant differences in the eclosion fraction (stream: p = 0.011; p = 0.332; p = 0.052) were observed based on stream, but not generation or experimental block (Figure S3). A post hoc analysis showed that stream B (eclosion fraction = 0.929) had a significantly higher eclosion fraction (p < 0.05) than stream A (eclosion fraction = 0.879) when pooling data from all generations and blocks. However, we note that the total sample size of all stream A and B samples throughout the experiment was high (N = 83). Therefore, it is important to compare the magnitude of these differences which shows that the eclosion fraction of stream B is only 5.73% higher than stream A. This relatively small increase in the context of seeing no significant differences in the total number of pupae and flies would suggest that changes in the eclosion fraction are unlikely to be part of a larger trend, especially when Figure S3 shows that most of this difference arose from Stream A – Generation 3 having a lower eclosion fraction than most treatments.

The data in the multigenerational reproductive assay was collected every 2 days over a period of 16 days allowing for the calculation of the mean emergence time of the pupae and flies. Significant differences in the mean pupation time (stream: p = 0.214; p = 7.55 × 10^–5^; p = 2.77 × 10^–3^) and mean eclosion time (stream: p = 0.104; p = 9.66 × 10^–7^; p = 7.44 × 10^–3^) were observed on the basis of generation and experimental block, but not stream (Figure c,d). Post hoc analysis revealed that these differences typically arose from generation 1 having a faster pupation and eclosion time than the other generations.

Few studies exist examining the effect of microplastics on Drosophila melanogaster over multiple generations. Jimenez-Guri et al. exposed Drosophila melanogaster to control, 1% polyethylene microplastics, and 1% polyvinyl chloride (PVC) microplastics food during the first generation, but only control food in the second generation. In the first generation, no significant differences in the number of pupae, number of flies, mean pupation time, and mean eclosion time between treatments was observed. However, in the second generation, Drosophila melanogaster, with parents raised on polyethylene or PVC in the first generation, had significantly shorter pupal stages than those with parents raised on control food. It was also determined that the number of flies in the second generation with parents raised on polyethylene was 76% higher than those with parents raised on the control. While their study showed a similar lack of any toxicity in the first generation, some effects were observed in the second generation. This may be due to differences in the size and concentration of the exposed particles. In the present study, microplastics that are <38 μm were used whereas Jimenez-Guri et al. used 23–500 μm particles, with 90% being >125 μm in size. Likewise, they used a concentration of 1% polyethylene corresponding to approximately 10 000 mg/kg compared to the present study which used a concentration of 100 mg/kg. Another study by Liu et al. found that exposing Drosophila melanogaster parental flies to 0.1 μm amino-functionalized polystyrene particles at a concentration of 50 μg/L resulted in female flies laying fewer embryos. However, no significant differences in the female or male fly mass of the offspring were observed compared to the control. A study involving the exposure of Drosophila melanogaster to 2 μm polyethylene terephthalate (PET) microplastics found that a 20 g/L exposure resulted in a significant 50% decrease in oviposition, whereas the decrease was not significant at 1 and 10 g/L. This shows the importance of covering a wide range of sizes and concentrations when examining toxicity throughout the literature.

HDPE microplastics were not found to affect the long-term reproductive performance of Drosophila melanogaster over multiple generations. Even when significant differences were observed, although small, it was a nontoxic response showing that pupae raised on HDPE microplastic-containing food for multiple generations had a higher eclosion fraction than those raised on CTRL food for multiple generations.

Sublethal Toxicity Assays

No negative impact of the HDPE microplastics at the concentrations tested were observed on the development of larvae into pupae and flies, or the reproductive performance of the flies over 4 generations. For a comprehensive investigation of toxicity, sublethal assays that were designed to detect toxicity at concentrations that do not induce death (i.e., sublethal) were performed. The locomotion of both larvae and adult flies and the mass of female and male flies were measured. Figure a shows that there were no significant differences in the number of contractions per minute of larvae based on treatment (p = 0.716) and experimental block (p = 0.795). Here, the treatment refers to the control (CTRL) or 100 mg/kg HDPE microplastics food. There were no significant differences between CTRL and HDPE-fed flies (p = 0.056) (Figure b) when measuring the fraction of flies which were able to climb 10 cm in 10 s or less, but there were differences between experimental blocks (p = 1.48 × 10^–3^). There were also no significant differences on the basis of treatment or experimental block, on the mass of female (treatment: p = 0.528; p = 0.304) (Figure c) or male (treatment: p = 0.488, p = 0.359) (Figure d) flies.

The sublethal toxicity of microplastics in Drosophila melanogaster has also been documented previously in the literature. A study whereby Drosophila melanogaster were exposed to 4, 10, or 20 μm polystyrene microplastics for 7 days in concentrations ranging from 2.604 to 2604 mg/kg food found that the percentage of flies able to climb 7 cm in 10 s decreased as concentration increased with all sizes. At the highest concentration, the climbing ability of the flies showed a significant decrease from 100% in the control to 51%, 56%, and 59%, when exposed to 4, 10, and 20 μm polystyrene microplastics, respectively. At 1302 mg/kg, a significant decrease for all three microplastics was still observed, but the result was no longer significant at lower concentrations. Similarly, the present study used <38 μm HDPE microplastics at a concentration of 100 mg/kg food and saw no impact on the climbing ability of the flies. However, in another study, flies were exposed to 0.1 and 1 μm polystyrene plastic beads at a concentration of 200 mg/L for 7 days before evaluating their climbing ability. Whereas 100% of both male and female flies passed the climbing test (i.e., ability to climb 9.5 cm within 120 s) in the control, only 80% and 66% of male and female flies, respectively, passed the test after exposure to the 0.1 μm polystyrene, and only 83% and 68% of male and female flies, respectively, passed the test after exposure to the 1 μm polystyrene. The size of these plastic particles was considerably smaller than our work, as well as the concentration being approximately double the 100 mg/kg we used. Polystyrene is also chemically different from polyethylene and could have different biological effects as a result.

No sublethal toxicity from the HDPE microplastics at a concentration of 100 mg/kg on the locomotion of larvae and flies, or the mass of female and male flies was observed in the present study. Similar to the outcome of the developmental toxicity assay, this may be a result of Drosophila melanogaster being unable to break down polyethylene or potentially avoiding the larger microplastics altogether due to the relative size of D. melanogaster both in the larval and adult fly form. This could render the polyethylene microplastics effectively inert, with the nanoplastic component being at too low a concentration to have any observable impact on the end points measured. The formation of eco-coronas on the surface of the nanoplastics may have also played a role in diminishing their impact. Although this study was conducted in a controlled environment rather than the natural environment, several biomolecules such as proteins, lipids, polysaccharides, and nucleic acids, that are excreted by Drosophila melanogaster can contribute to the formation of nanoparticle coronas. These coronas can then mask the underlying nanoplastic, a phenomenon which has been shown to reduce nanoplastic toxicity toward organisms.

Light Sheet Fluorescence Microscopy

To determine if the flies were able to uptake HDPE microplastics, on day 10, female flies that were raised on control and 10 000 mg/kg HDPE were sampled for light sheet imaging at a later date. Figure shows that there were no signs of internalization of the microplastics, whereby the majority of the red signal (representing dyed HDPE) was found on the surface of the flies (Figure b). From Figure , it was apparent that the microplastics have many different shapes and sizes, and the varying degree to which they may aggregate creates further polydispersity. On a mass basis, most of the particles were likely too large relative to the fly’s proboscis. Fruit flies ingest their food by secreting enzymes to break down solid food before consuming it in liquid form. , Therefore, it may be possible for the flies to avoid ingestion of the larger plastic particles if they cannot be broken down by their enzymes. Conversely, on a number basis, the sub-micrometer-sized particles may form the majority, but their mass concentration may have been too low to be detected. Ingestion of the sub-micrometer sized particles is more likely, since their small size may allow them to be more easily suspended in a liquefied form even if they cannot be broken down. It has also been shown that Drosophila melanogaster larvae can ingest and accumulate 20 nm and 1 μm polystyrene particles in their gastrointestinal tracts. These results are consistent with the overall lack of toxicity observed. Although surface contact of the HDPE microplastics occurred with the fruit fly, it was not a route of exposure capable of inducing observed toxicity in this case.

X-ray Microtomography

The fourth (i.e., final) generation of flies in each stream were imaged using μCT (Figure S4). Some of the treatment specimens had severe malformations. BF4–1 (i.e., stream B, generation F4, replicate id 1, in Figure S4) had an expansive ventrally located thoracic cavity that displaced the anterior midgut dorsally and the ventral nerve cord to a location right of the midline. All three CF4 samples had malformations of their salivary glands which were either absent, crossed the midline, or had malformed expansions in the abdomen. BF4–1 and all three CF4 specimens had malformed scutella. The scutellum is a dorsal process projecting from thoracic segment two and houses a complex pumping and valve organ to pump hemolymph throughout the wings. Unfortunately, wings were not imaged in this study. However, recently Sorensen et al. discovered geometric variations in wing size and shape in Drosophila melanogaster exposed to polystyrene microplastics and nanoplastics. Although they did not examine the scutellum, we propose that this organ and the wings may be viable anatomical markers of sublethal effects of microplastics on Drosophila melanogaster for future research. Both structures are derived from imaginal wing discs, which are model developmental systems. The well characterized cellular and genetic process of these imaginal discs may provide fruitful resources to identify sensitive, nonlethal ecotoxicological disruptions of their developmental pathways. The scutellum and wings are easily imaged using standard light microscopy and can provide efficient, high throughput assays. Furthermore, flight performance assays may provide insight into the functional consequences of these sublethal effects.

A previous study by Matthews et al. similarly demonstrated tissue damage in adult female Drosophila melanogaster after exposure to 20 nm and 1 μm dialyzed fluorescent polystyrene nanoplastics and microplastics, respectively. They also found that there was a 4.5× higher expression of the Hsp70 (heat shock proteins) gene in the flies exposed to 10 ppm microplastics compared to the control. Similarly, the endocrine disrupting potential of polyethylene microplastics has previously been documented in fish (Oryzias latipes). Future studies should be performed to monitor the gene expression of hormones critical to the development of adult flies. Zhang et al. previously demonstrated that 0.1 and 1 μm polystyrene nanoplastics and microplastics, respectively, induced intestinal cell damage in third instar Drosophila melanogaster larvae. Our work further demonstrates the impact of long-term multigenerational microplastic exposure using polyethylene and artificial weathering to simulate microplastics as they may be found in the natural environment.

Conclusions

Several assays were performed to measure the toxicity of HDPE microplastics (<38 μm) in the fruit fly, Drosophila melanogaster. The effect of the microplastics on the development of larvae into adult flies was measured in concentrations ranging from 0.1 to 10 000 mg/kg HDPE in food. No toxicity was observed in the number of pupae and flies that emerged, the eclosion fraction, or the mean pupation and eclosion time. Further assays were performed at 100 mg/kg HDPE to be more representative of HDPE concentrations found in the environment. The effects of chronic HDPE microplastic exposure on the reproductive performance of flies over four generations were evaluated. The presence of HDPE microplastics in the fly food had no effect on reproductive performance over the four generations observed in terms of the total number of pupae and flies that emerged, and the mean pupation and eclosion time. However, the eclosion fraction was 5.7% higher (p < 0.05) in the stream of Drosophila melanogaster that were fed HDPE microplastics throughout all generations when compared to those that were always fed control food. To ensure that we did not miss certain sublethal toxic effects (e.g., toxicity that is not severe enough to lead to a decline in the number of pupae and flies), the effect of the HDPE microplastics on the locomotion of larvae and adult flies and the mass of female and male flies was measured. No sublethal toxicity was observed on these end points after exposure to HDPE microplastics. Overall, the lack of toxicity is not unusual. For instance, Liang et al. exposed Drosophila melanogaster to 1 g/L polyethylene terephthalate microplastics (2 μm) and found that it lengthened the lifespan of male flies, suggesting that not all microplastic effects on organisms are necessarily toxic.

The results of the present study paint a comprehensive picture of the lack of toxicity from HDPE microplastics on Drosophila melanogaster. A wide range of concentrations in the developmental toxicity assay were used to determine acute mortality. Further tests were performed at a more environmentally relevant concentration of 100 mg/kg HDPE in food to assess multigenerational and sublethal effects. This work found that no significant toxicity was observed from the HDPE microplastics on Drosophila melanogaster, however malformations were observed and may warrant further investigation.

Supplementary Material