Lipidomic alterations in oysters caused by environmentally relevant exposure to microplastics and estrogenic endocrine disrupting chemicals

TL;DR

This study shows how microplastics and estrogen-like chemicals affect oyster lipid levels, with estrogens having a stronger impact than microplastics.

Contribution

The study reveals the differential and combined effects of microplastics and estrogens on oyster lipidomes at environmentally relevant concentrations.

Findings

Estrogens had a stronger effect on oyster lipidomes than microplastics.

Oyster digestive glands showed a stronger response to stress compared to other tissues.

Combined exposure to microplastics and estrogens did not cause greater stress than individual exposure.

Abstract

Microplastics and estrogens affect oysters differently.

Estrogens are more bioavailable than microplastics in oysters.

Amplification of EEDC effects by microplastics was low in molluscs.

Incubation of EEDCs with smooth surfaced PE-MPs did not increase adsorption.

Oysters showed different adaptive responses at low MPs and EEDCs exposure.

The online version contains supplementary material available at 10.1007/s10646-026-03055-2.

There is a ubiquitous co-occurrence of microplastics (MPs) and estrogenic endocrine disrupting chemicals (EEDCs) in aquatic environments. Their combined presence may induce sub-lethal stress in aquatic organisms, like molluscs, which may adapt to the stress through adjusting the lipidome. This study explored the individual and combined effects of polyethylene microplastics (PE-MPs) and a mixture of EEDCs at environmentally realistic concentrations on the lipidome of the Sydney rock oyster (Saccostrea glomerata). After seven days exposure, lipids were measured in the digestive glands and gonads using proton nuclear magnetic resonance (1H NMR) spectroscopy. Permutational analysis of variance (PERMANOVA) revealed the lipidome of oysters exhibited treatment differences with interactions between sex and tissue type. Subsequent post-hoc analysis revealed the male digestive gland to be significantly more responsive than the other tissues tested. An increase in cholesterols and aldehydes in all treatments compared to controls was ascribed to an adaptive response to transient oxidative stress in oysters, which was alleviated by increased food intake. There was a significantly greater effect of the EEDCs mixture than PE-MPs in oysters. The combined oxidative stress of PE-MPs and EEDCs from co-exposure did not exceed the individual effects of PE-MPs and EEDCs, which may relate to low exposure concentrations of PE-MPs, selective rejection of MPs in pseudofeces, or poor vector role of smooth surfaced PE-MPs.

The online version contains supplementary material available at 10.1007/s10646-026-03055-2.

Figures

Microplastic uptake rate of Sydney rock oysters (particles/oyster/day), *S. glomerata* (*n* = 6 per treatment). Different letters denote statistically significant treatment differences based on one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) test. Box is median with 25^th^ and 75^th^ percentile, line inside the box indicates the median, whiskers extend from the box to the minimum and maximum data points, and dots show the distribution of individual data points

Lipidome alterations in the digestive gland of female Sydney rock oysters, *Saccostrea glomerata* (n = 7 per treatment). Different letters denote statistically significant treatment differences based on one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) test. Box is median with 25^th^ and 75^th^ percentiles, line inside the box indicates the median, whiskers extend from the box to the minimum and maximum data points, and dots show the distribution of individual data points.

Lipidic metabolite alterations in the digestive gland of male Sydney rock oysters, *Saccostrea glomerata* (*n* = 7 per treatment). Different letters denote statistically significant treatment differences among treatments via one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) test. Box is median with 25^th^ and 75^th^ percentile, line inside the box indicates the median, whiskers extend from the box to the minimum and maximum data points, and dots show the distribution of individual data points

Lipidic metabolite alterations in the gonad of female Sydney rock oysters, *Saccostrea glomerata* (n = 7 per treatment). Different letters denote statistically significant differences among treatments via one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) test. Box is median with 25^th^ and 75^th^ percentiles, line inside the box indicates the median, whiskers extend from the box to the minimum and maximum data points, and dots show the distribution of individual data points

Full text

Introduction

The ubiquitous presence of MPs and endocrine disrupting chemicals (EDCs) in aquatic environments is well known (Fu et al. 2024). A subset of EDCs are the estrogenic EDCs (EEDCs), like estrone (E1), 17β-estradiol (E2), and estriol (E3), 17α-ethinyl estradiol (EE2), bisphenol A (BPA), nonyl phenol (NP), and octyl phenol (OP). Both MPs and EEDCs find their way into aquatic environments, primarily through anthropogenic activities like municipal, industrial, hospital, and wastewater effluents (Rossatto et al. 2023; Grzegorzek et al. 2024). The presence of these contaminants in aquatic environments has resulted in a wide range of documented adverse effects in aquatic organisms, including oxidative stress and perturbations to reproductive endpoints in bivalves (Andrew et al. 2008; Tran et al. 2016; Mai et al. 2023). Because bivalves are generally sedentary and spend most of their lifetime in the same region, they are prone to exposure to these pollutants (Andrew et al. 2008; Lee et al. 2016), including the risk of co-exposure.

Although very different, MPs and EEDCs share similar mechanistic pathways of toxicological effects, including being known to generate oxidative stress and reactive oxygen species (ROS) upon accumulation in molluscs (Islam et al. 2022; Mai et al. 2023). Lysosome bound MPs after accumulation are targets of the innate immune system which generate ROS (Von Moos et al. 2012), while estrogens are metabolised to quinones generating ROS in the process (Livingstone et al. 1990). ROS such as the superoxide ion and peroxides are quenched via antioxidant systems such as catalase (CAT), superoxide dismutase (SOD), glutathione peroxidase (GPx) and glutathione S-transferases (GSTs). When the capacity of antioxidant systems is exceeded, ROS can oxidise lipids leading to the production of lipid peroxides and aldehydes (Alves de Almeida et al. 2007; Ayala et al. 2014). For example, MP exposure increased diradylglycero- lipids (46:0 and 49:1), while reducing triradylglycero lipids (56:1) in the digestive glands of mussels (Moncrieffe et al. 2023). Similarly, estrogenic mixtures significantly affected lipid metabolism by forming lipid peroxides due to a transient increase in oxidative stress in mussels (Canesi et al. 2008). A decrease in essential and long-chain fatty acids was similarly observed in unionid mussels exposed to EE2, with a general increase in lysolipids (Leonard et al. 2014). In oysters, a mixture of EEDCs significantly decreased phospholipids and phosphatidylcholine (Islam et al. 2022). The reduction of these lipids might have occurred due to lipid peroxidation through estrogen metabolism, reduction in substrate availability for lipid synthesis, or a reduction in algal consumption that provides the building blocks for these molecules (Islam et al. 2022).

Lipids are of special interest since they play several important biological functions, including contributing to the structural integrity of cell membranes, and acting as signalling molecules in cellular homeostasis regulation (Parton and Simons 2024). Alterations in lipids affect spat production, survival during starvation, ability to respond to the surrounding environment, and gonadal development in molluscs (Gallager et al. 1986; Laudicella et al. 2020a, b), which finally impact ecosystem health and aquaculture sustainability. To the best of our knowledge, no previous studies have explored the effects of co-exposure to MPs and EEDCs on the lipidome in molluscs, however, this may be of particular relevance since MPs and EEDCs are both largely hydrophobic (Guedes-Alonso et al. 2021). The hydrophobicity of both contaminants is expected to make MPs an efficient carrier of organic contaminants via adsorption-desorption mechanisms on their surface. For example, hydrophobic EE2 was easily adsorbed onto MPs up to concentrations as high as 1654 µg/g in seawater, when EE2 and MPs concentrations were 10 µg/L and 0.05 g/L respectively (Lu et al. 2021). The concentration of adsorbed organic contaminants on MPs becomes more pronounced after aging the MPs (Bhagat et al. 2022). The adsorption of hydrophobic organic contaminants may subsequently enhance uptake by aquatic organisms and elevate the bioavailability of these chemicals via ingestion (Avio et al. 2015; Pittura et al. 2018; Webb et al. 2020).

Within the same experimental framework, our previous metabolomics-based study, revealed sex and tissue specific metabolite alteration, particularly of polar metabolites including amino acids, carbohydrates and intermediates of the Kreb’s cycle (Kumar et al. 2024). The study indicated a possibility of stimulatory and/or a potential hormetic response in oysters due to single and mixed exposure of EEDCs and MPs (Kumar et al. 2024).

Therefore, this study aimed to specify the stimulatory and/or a potential hormetic effects of MPs and a mixture of EEDCs at environmentally relevant concentrations on the lipidome of oysters. Additionally, based on lipidomics, the mechanistic insights of the effects, dietary influences on effect mitigation, and the vector role of MPs were evaluated. First, the differential effects of MPs versus a mixture of EEDCs was evaluated, hypothesizing that EEDCs would have more pronounced effects on the lipidome than MPs given their relatively elevated transport to organs. Second, co-exposure effects were evaluated through lipidomic changes, while hypothesizing that mixture effects would be greater than single exposures as MPs may act as a vehicle of enhanced EEDCs uptake by oysters. Third, we explored whether pre-incubation or simultaneous exposure of MPs with EEDCs enhances the negative effects on oysters. It was hypothesized that pre-incubation would result in more EEDCs being adsorbed on MPs and would increase EEDCs exposure relative to simultaneous addition and exposure.

Materials and methods

Experimental design

The experimental design follows a protocol previously described in Kumar et al. (2024). Briefly, mature oysters (age: 2.5 years) were sourced from Commarty Bay, Port Stephens. Acclimatisation of oysters was performed for 7d in filtered seawater (salinity: 33–35) at a temperature of 21 ± 1 °C, at the NSW Department of Primary Industries, Port Stephens Fisheries Centre (O’Connor et al. 2008). During acclimatisation and exposure, oysters were fed a mixture of algal species (Chaetoceros meulleri,* Tisochrysis lutea*, and Diacronema lutheri at a ratio of 2:1:1) at a rate of 2 × 10^9^ cells/oyster/day.

For exposure of MPs and EEDCs, there were five treatments: a solvent control, PE-MPs, EEDCs, pre-incubated PE-MPs/EEDCs, and simultaneous PE-MPs/EEDCs. Each treatment had seven replicate aquaria with four oysters in each (i.e., a total of 28 oysters in each treatment, or N = 7 independent replicates per treatment). No oyster mortality was observed during the exposure period. Exposures were run for 7 d. Every day (after 24 h of exposure), all experimental aquaria were cleaned, seawater was renewed, and respective PE-MPs and/or EEDCs doses were reapplied in the treatments.

The control treatment incorporated 0.0002% dimethyl sulfoxide (DMSO) to dissolve EEDCs. In the PE-MPs treatment, there was 0.55 µg/L or approximately 66 particles/L of polyethylene microplastics (PE-MPs) along with 0.0002% of DMSO. The yellow-coloured PE-MPs had a size range of 10–45 μm and a density of 1 g/cc (Cospheric LLC, California, USA). The EEDC treatments contained E1, E2, E3, EE2, BPA, 4-t-OP and 4-NP at 20, 10, 5, 5, 500, 500, and 5000 ng/L, respectively, plus 0.0002% DMSO. These concentrations of EEDCs were chosen as being typical of effluents and/or receiving waters in Australian estuaries (Tran et al. 2019). In the pre-incubated PE-MPs/EEDCs and simultaneous PE-MPs/EEDCs treatments, oysters were exposed to 0.55 µg/L of PE-MPs, EEDCs at the previous concentrations, and 0.0002% DMSO. In the pre-incubation treatment, PE-MPs and EEDCs were combined for 48 h prior to exposure, whereas in the simultaneous treatment, both were introduced together at the start of the exposure without pre-incubation.

Nominal concentrations of EEDCs and PE-MPs after spiking and before water changes were verified. PE-MP accumulation after seven days of exposure was verified by visual examination of the oyster’s gut. Briefly, EEDCs from the water were extracted using a solid phase extraction technique coupled with Visiprep™ SPE vacuum manifold system (Sigma-Aldrich, Inc., Missouri, USA). The EEDC concentrations in water were estimated using liquid chromatography-mass spectroscopy (LC-MS) (Nexera X2 UPLC coupled to a SCIEX 6500 + QTRAP MS, Shimadzu, Tokyo, Japan). MPs were filtered on glass fiber filters after preliminary clean up, including digestion of samples by potassium hydroxide. Identification and quantification of MPs were performed using a stereomicroscope (Olympus SZ61) with a magnification of 20 × (Olympus, Tokyo, Japan). Detailed procedures for EEDC and MP quantification can be found in Kumar et al. (2024), and data related to EEDC and MP concentrations for this study are provided in supplementary Tables S1-S2.

Sex identification, tissue collection, and lipid extraction

Sex identification was performed during tissue collection by examining gonad samples under a stereo microscope for the presence of eggs or spermatozoa. After dissection, the digestive glands of both sexes and gonad tissue from each female oyster (50–70 mg/replicate) were individually collected and stored in pre-cleaned Eppendorf tubes. The digestive gland was selected because of its fundamental role in food digestion and metabolic transformation of contaminants, whereas the female gonad was specifically chosen based on previous studies demonstrating its sensitivity to estrogenic mixtures (Islam et al. 2021, 2022). Tissues were instantly frozen with liquid nitrogen and stored at -80 °C until lipid extraction.

Lipids were extracted using the modified lipid extraction methods described by Bligh and Dyer (1959). Briefly, the pooled frozen tissue samples of four oysters from replicate aquaria (n = 7 for each treatment) were thawed and polar and non-polar metabolites were extracted using methanol (CH_3_OH), chloroform (CHCl_3_) and deuterated water (dH_2_O). First, 400 µL ice-cold MeOH was added to thawed tissues and homogenised for 1–2 min with a tissue homogeniser (TissueRuptor, Qiagen Inc., Hilden, Germany). Second, the CH_3_OH homogenate was incubated overnight at -20 °C. The next day, 800 µL CHCl_3_ and 320 µL dH_2_O were added and the homogenate was vortexed for 30 s (Zhu et al. 2015). The resultant solution was then separated into two layers by centrifugation for 10 min at 16,000× g with a controlled temperature at 4 °C. The bottom CHCl_3_ layer containing the lipidome was carefully partitioned into a glass vial. The remaining protein pellets were extracted a second time using the same protocol to ensure maximum recovery of lipids. The combined extracts (about 1600 µL) from the repeated extraction were preserved at − 80 °C until further processing (Beckonert et al. 2010). All experimental procedures complied with institutional and national guidelines for the use of invertebrates in research, and ethical approval was not required for oysters under Australian regulations.

Lipid extract preprocessing and 1H NMR analysis

Lipid analysis in ^1^H NMR requires a preprocessing of the extract. First, it requires drying in a Series II centrifugal vacuum concentrator (GeneVac Technologies, Ipswich, England), followed by freeze-drying to remove any residual water. Second, suspension of the dried lipidic extract in 500 µL deuterated chloroform (CDCl_3_) containing 0.05% sodium-3-(tri-methylsilyl)-2, 2, 3, 3- tetradeuteriopropionate (TSP) as an internal reference was undertaken. The CDCl_3_ was used for field locking purposes. Third, removal of any debris through vortex and centrifugation was performed (5 min, 2500×g, at 4 °C). Finally, purified extracts were transferred to 5 mm NMR tubes for subsequent proton nuclear magnetic resonance (^1^H NMR) spectroscopy (Beckonert et al. 2010; Melvin et al. 2018).

The ^1^H NMR spectroscopic analysis was performed in 800 MHz Brucker^®^-Advance III HDX spectrometer, equipped with a Triple Resonance 5 mm cryoprobe with Z-gradient. The proton spectra were acquired with the zg30 pulse program along with 128 scans, relaxation delay of 1.0 s, pulse width of 8.20 µs, and spectral width of 16 kHz. Then, the edited ^1^H^13^C heteronuclear single quantum coherence (HSQC) spectra were obtained using 100 scans, relaxation delay of 0.8 s, pulse width of 7.85 µs, and spectral widths of 12.8 and 33.1 kHz. The resultant spectra were used to assign peaks to characteristic functional groups and identify the corresponding lipids integrals using the Mnova software (MestReNova v14.3.1, Santiago de Compostela, Spain). Lipids were identified from an in-house library developed through previous studies and cross-referenced with the published literature as per Melvin et al. (2019).

Data pre-treatment and statistical analysis

The pre-treatment of lipidic data involved four consecutive steps before statistical analysis. First, extracted lipid integrals were normalized by tissue weight for each replicate in each treatment. Second, the 26 lipid integrals were sorted and averaged to 8 major lipid classes (e.g., to phospholipids, glycolipids, triacylglycerol, fatty acids, and cholesterols) and 1 lipid breakdown product (aldehydes). Third, the relative lipid concentration data were checked for outliers among the replicates in each treatment and any outliers > 2 standard deviations from the mean were removed. Fourth, relative lipids concentrations were pareto scaled and log_10_ transformed for subsequent statistical analysis.

The statistical analyses involved permutational analysis of variance (PERMANOVA) and one-way analysis of variance (ANOVA) tests. These tests were performed using Primer 7 with PERMANOVA add-on (Massey University, Auckland, New Zealand), SPSS 29 (SPSS Inc., Chicago, Illinois, USA), and graphs were prepared using GraphPad Prism 10.4 (GraphPad Software, Boston, USA).

• Lipids were analysed using a two-way PERMANOVA to evaluate the effects of tissue and treatment on the lipidome. There were two factors: tissue (female digestive gland, male digestive gland, and female gonad) and treatment (control, PE-MPs, EEDCs, pre-incubated PE-MPs/EEDCs, and simultaneous PE-MPs/EEDCs). Pairwise post-hoc tests were applied to understand the main effect of treatment and tissue differences. When the treatment by tissue interaction was significant (p < 0.05), a pairwise post-hoc test was applied to identify specific treatments that significantly varied in each type of tissue.
• Tissue specific individual metabolite group differences among the five treatments were determined by a one-way ANOVA. Tukey’s honestly significant difference (HSD) test was performed to determine significance levels of metabolites among treatments. A p value less than 0.05 was considered a statistically significant difference. The one-way ANOVA was performed to identify which specific lipids varied significantly, and which tissues and treatments influenced the variance in the data.

Results

MPs in exposure media and uptake by oysters

PE-MPs were detected in the exposure water across all MP-containing treatments immediately after spiking and prior to water renewal at 24 h (Table S2). No significant differences in water-phase PE-MPs concentrations were observed among MP treatments at either time point. MP concentrations in the water were higher prior to water change than immediately after spiking, indicating redistribution or enhanced suspension of particles during exposure.

MPs were quantified in gut samples collected at the end of the exposure period and the uptake rate are presented in Fig. 1. MPs were detected in the gut of oysters exposed to only MP-containing treatments. Uptake rates were highest in the PE-MP treatment (average: 1.93, range: 0.71–4.57 particles/oyster/day), with greater variability among individuals compared to co-exposure treatments (Fig. 1). However, no statistically significant differences in MP uptake rate were observed among PE-MP treatments (F = 1.37, p = 0.33). Although MP uptake rates did not differ significantly among treatments, oysters exposed to PE-MPs alone exhibited higher median uptake and greater inter-individual variability. It may indicate that differences in observed effects among treatments are unlikely to be driven solely by differences in MP ingestion rate but may instead reflect variations in PE-MPs/EEDCs interactions or bioavailability.

Fig. 1. Microplastic uptake rate of Sydney rock oysters (particles/oyster/day), S. glomerata (n = 6 per treatment). Different letters denote statistically significant treatment differences based on one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) test. Box is median with 25^th^ and 75^th^ percentile, line inside the box indicates the median, whiskers extend from the box to the minimum and maximum data points, and dots show the distribution of individual data points

Tissue and treatment effects on the lipidome

A two-way PERMANOVA indicated significant effects of treatment (Pseudo-F = 5.60, p = 0.001), tissue (Pseudo-F = 14.1, p = 0.001), and their interaction (Pseudo-F = 3.74, p = 0.001) (Table 1). This suggests that the effects of treatment on lipids varies with tissue type.

Table 1. Two-way permutational multivariate analysis of variance (PERMANOVA) of relative lipid abundance among treatments (i.e., control, PE-MPs, EEDCs, Pre-incubated PE-MPs/EEDCs, and simultaneous PE-MPs/EEDCs) and tissues (i.e., female digestive gland, male digestive gland, and female gonad)SourcedfSSMSPseudo-Fp (perm)Unique permsTreatment4.004.481.125.600.001999Tissue2.005.592.8014.10.001998Treatment × Tissue8.005.970.753.740.001999Residual90.0018.10.20Total104.0034.0PE-MPs indicates polyethylene microplastics and EEDCs indicates estrogenic endocrine disrupting chemicals

Since the treatment versus tissue interaction was significant (p = 0.001), a post-hoc pairwise comparison for each tissue type was performed. The post-hoc test revealed that the male digestive gland is the most sensitive tissue to both types of contaminants exposure. In this tissue, lipids in all treatments showed significant differences compared to the control (Table 2). Further, the effect of each contaminant was unique, with PE-MPs and EEDCs exhibiting significantly different resulting lipidomes (p = 0.002). There was a mixture effect, with the combined simultaneous treatment lipidome varying significantly from individual contaminant treatments (p = 0.001). Finally, pre-incubation of PE-MPs with EEDCs resulted in a significantly different lipidome compared to simultaneous exposure to these contaminants (p = 0.001).

The female gonad was less sensitive than the male digestive gland. In the female gonad, lipids in the PE-MPs and pre-incubated PE-MPs/EEDCs treatments differed significantly from the control. Also, mixture effects were significantly different in the lipidomic assemblages compared to the PE-MPs treatment alone (Table 2).

The female digestive gland was the least responsive tissue. In this tissue, lipids in the PE-MPs treatment and the pre-incubated PE-MPs/EEDCs treatment varied significantly from the control. Like male digestive gland, the effects of PE-MPs and EEDCs on lipids were significantly different (p = 0.025). (Table 2).

Table 2. Tissue-specific pair-wise permutational multivariate analysis of variance (PERMANOVA) of relative lipids abundances among treatments in tissues (i.e., female digestive gland, male digestive gland, and female gonad) based on Euclidean distance-based resemblance matrixTreatmentsFemale digestive glandMale digestive glandFemale gonadtp (perm)unique permstp (perm)unique permstp (perm)unique permsControl, PE-MPs2.59 0.014 7533.47 0.001 7523.57 0.007 779Control, EEDCs1.900.0567612.25 0.003 7621.500.154754Control, Pre-incubated PE-MPs/EEDCs2.21 0.025 7552.64 0.001 7622.01 0.034 764Control, Simultaneous PE-MPs/EEDCs1.680.0737573.93 0.003 7672.020.037770PE-MPs, EEDCs2.51 0.025 7652.69 0.002 7541.600.142760PE-MPs, Pre-incubated PE-MPs/EEDCs1.620.1227651.580.1657392.19 0.020 753PE-MPs, Simultaneous PE-MPs/EEDCs1.480.1357534.75 0.001 7491.84 0.041 757EEDCs, Pre-incubated PE-MPs/EEDCs1.110.2837461.280.2137600.640.654752EEDCs, Simultaneous PE-MPs/EEDCs1.540.1157594.22 0.001 7560.940.458773Pre-incubated PE-MPs/EEDCs, Simultaneous PE-MPs/EEDCs0.990.4227703.91 0.001 7480.710.656776Bold values indicate the significant results

Tissue specific alteration in lipids

Female digestive gland

In the female digestive gland, only three lipid classes (i.e., phospholipids, esterified cholesterols, and total cholesterols) and aldehydes showed significant treatment differences (Fig. 2). Compared to controls, phospholipids increased in individual and mixed treatments, where a significant increase was observed only in the PE-MPs and pre-incubated PE-MPs/EEDCs treatments. No significant treatment differences were observed among single and mixed treatments (Fig. 2a). For esterified and total cholesterols, a significant reduction was observed in EEDC exposed oysters compared to control. No difference among individual PE-MPs and mixed PE-MPs treatments, nor between mixed treatments were observed (Fig. 2f and h). Aldehydes showed a significant treatment difference (p < 0.001), where single and mixed treatments significantly differed from control. Neither single nor mixed treatments varied significantly from each other (Fig. 2i).

Fig. 2. Lipidome alterations in the digestive gland of female Sydney rock oysters, Saccostrea glomerata (n = 7 per treatment). Different letters denote statistically significant treatment differences based on one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) test. Box is median with 25^th^ and 75^th^ percentiles, line inside the box indicates the median, whiskers extend from the box to the minimum and maximum data points, and dots show the distribution of individual data points.

Male digestive gland

The male digestive gland was more responsive than the female digestive gland, showing a significant treatment difference in seven out of eight lipid classes and aldehydes (Fig. 3). Compared to the control, esterified cholesterol and total cholesterol showed a significant increase in single and mixed PE-MPs treatments. These lipids were also significantly higher in PE-MPs than EEDCs treatment, but no significant variation was observed between mixed treatments.

Glycolipids showed a significant increase in simultaneous PE-MPs/EEDCs treatments compared to the control. Neither individual nor mixed treatments showed a significant difference with each other (Fig. 3b).

Triacylglycerol showed significant increases in simultaneous PE-MPs/EEDCs, while a significant reduction was observed in pre-incubated PE-MPs/EEDCs treatments compared to the control. Single treatments did not vary significantly from the control treatment. The two mixed treatments varied significantly, where there was significantly higher triacylglycerols in simultaneous PE-MPs/EEDCs treatment compared to the pre-incubated PE-MPs/EEDCs treatment. Compared to single treatments, triacylglycerols significantly increased in simultaneous PE-MPs/EEDCs (Fig. 3c).

Although aldehydes indicated an overall treatment difference (p < 0.001), none of the treatments varied significantly from the control treatment (Fig. 3i).

Fig. 3. Lipidic metabolite alterations in the digestive gland of male Sydney rock oysters, Saccostrea glomerata (n = 7 per treatment). Different letters denote statistically significant treatment differences among treatments via one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) test. Box is median with 25^th^ and 75^th^ percentile, line inside the box indicates the median, whiskers extend from the box to the minimum and maximum data points, and dots show the distribution of individual data points

Female gonad

In female gonad, fewer lipid classes were significantly different among treatments than the male digestive gland, but there was a higher induction rate than the female digestive gland. Six out of eight lipids and aldehydes showed significant treatment differences (Fig. 4).

Phospholipids showed a treatment difference, but none of the treatments significantly varied from the control (Fig. 4a). Aldehydes showed significant treatment differences, where they increased only in the EEDCs treatment compared to the control (Fig. 4i).

Triacylglycerol, total fatty acids, and esterified cholesterols showed treatment differences, where only the PE-MPs treatment significantly differed from the control (Fig. 4a). Triacylglycerol and total fatty acids significantly increased in PE-MPs treatment compared to control, while esterified cholesterols were significantly reduced in PE-MPs treatment compared to control (Fig. 4c and e, and 4f).

A significant treatment difference was observed for total cholesterols. Results exhibited significant decreases in total cholesterols in the PE-MPs treatment, while significant increases were observed in the EEDCs treatment relative to the control. Mixed treatments did not show significant variability between themselves nor from controls. Total cholesterol significantly increased in the simultaneous PE-MPs/EEDCs treatment when compared to PE-MPs treatment. However, total cholesterol significantly decreased in simultaneous PE-MPs/EEDCs treatment when compared to EEDCs treatment (Fig. 4h).

Fig. 4. Lipidic metabolite alterations in the gonad of female Sydney rock oysters, Saccostrea glomerata (n = 7 per treatment). Different letters denote statistically significant differences among treatments via one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) test. Box is median with 25^th^ and 75^th^ percentiles, line inside the box indicates the median, whiskers extend from the box to the minimum and maximum data points, and dots show the distribution of individual data points

Discussion

In this study, single and co-exposure of PE-MPs with a mixture of EEDCs at environmentally relevant doses did not show a strong, nor consistent alteration in the oysters’ lipidome. But lipidome alteration in oysters among treatments varied with tissue type. Among the three tissues, the lipidome in the male digestive gland responded more than the female digestive gland or gonad. This suggests that the lipid profiles of male digestive gland may be more sensitive to these contaminants and, therefore, the most suitable for biomonitoring of environmentally relevant exposures of PE-MPs, EEDCs, and their mixtures. Similarly, not all lipids showed significant variation across treatments. In all tissue types, significant and consistent treatments differences were found for cholesterols and aldehydes. Thus, the alteration in cholesterols and aldehydes could be used as bioindicators of MP and/or EEDC exposure in molluscs.

Microplastics and estrogens induce transient oxidative stress

MPs and EEDCs induce oxidative stress in molluscs (Islam et al. 2022; Mai et al. 2023), generally characterised by an imbalance between cellular antioxidant defences and overproduction of free radicals, particularly the ROS. Subsequently, the overproduced ROS attack lipids, resulting in the formation peroxidation products, like aldehydes (Liu et al. 2024). Further, compensatory mechanisms to alleviate peroxidation effects includes the upregulation of phospholipid production. The literature also indicates that MP exposure in molluscs increases ROS production, lipid peroxidation, and aldehyde concentration (Webb et al. 2020; Li et al. 2022 Kılıç et al. 2023; Liu et al. 2024). Similarly, exposure of estrogens in molluscs results in increased ROS generation, lipid peroxidation, and potentially the generation of aldehydes as peroxidation products (Koutsogiannaki et al. 2014; Islam et al. 2022). In this study, the increases in aldehydes in exposed groups, especially for EEDCs compared to the control treatment indicates that these contaminants may have indeed caused transient oxidative stress in oysters. However, there was a difference in aldehyde concentration between individual PE-MP and EEDC treatments, which might indicate a differential severity or mode of oxidative stress induction in oysters. Probably, MPs are neutralized primarily through phagocytic mechanisms as foreign particulates, whereas EEDCs are metabolically detoxified via enzymatic biotransformation in molluscs.

MP ingestion can decrease the number of contacts between adjacent filaments, thicken and disorganise gill epithelia (Bråte et al. 2018; Sendra et al. 2021). MPs can also cause destabilisation of lysosomal membranes (Von Moos et al. 2012), and atrophy and disorder in digestive tubules (Mai et al. 2023). For example, PE-MPs (size: < 0–80 μm) ingested by mussels travelled from the gills to the digestive gland, where they accumulated in the lysosomal systems in a phagocytic process (Von Moos et al. 2012). When MPs are in phagocytes, the innate immune system is triggered to neutralize the foreign substance and high quantities of ROS are produced (Détrée and Gallardo-Escárate 2018; Sun et al. 2024).

In contrast to MPs, EEDCs might have induced oxidative stress through ROS production during biotransformation of xenobiotics (i.e., EEDCs). The ROS production arises from a redox cycling reaction, where xenobiotics are reduced in the presence of microsomal NADPH or NADH dependent flavoprotein reductase enzymes (Kappus 1986). Then, in the presence of oxygen, the reduced xenobiotics are oxidized by oxygen and oxygen is reduced to form superoxide anion radicals (O_2_¯·), and the newly formed xenobiotic radical is reduced back to the parent compound, completing a redox cycle (Livingstone et al. 1990). The superoxide anion radicals (O_2_¯·) may form hydrogen peroxide (H_2_O_2_) and subsequently transform into hydroxy radicals (OH·). Although the presence of estrogen receptors in molluscs is controversial, molluscs can metabolise estrogens (Balbi et al. 2019). Most likely, molluscs follow similar metabolic reactions found in vertebrate models, but through non-genomic pathways (Balbi et al. 2021). For example, common estrogens like estrone, estradiol, and ethynylestradiol are reduced to estrone-2,3-o-quinone or estrone-3,4-o-quinone, estradiol-2,3-o-quinone or estradiol-3,4-o-quinone, and ethynylestradiol-2 or 4 hydroxy- o-quinone, respectively. The reduction of the estrogens is accompanied by the production of ROS (Stack et al. 1996; Bolton and Dunlap 2017).

Regardless of the ROS generation pathways arising from MP and EEDC exposure, the ROS then attack the lipids in oysters resulting in an increase in aldehydes. The most profound ROS that attack lipids are hydroxyl (OH·) and hydroperoxyl (OH_2_·) radicals. The hydroxyl (OH·) radicals are highly mobile, water-soluble, and chemically reactive species that can react with any lipids located less than a nanometre from the origin of OH· in cells (Halliwell and Gutteridge 1984; Ayala et al. 2014). The rapid generation of ROS and lipid peroxidation in molluscs are well documented even at tissues distant from the exposed tissue. In hemocytes of the Mediterranean mussel, E2 exposure significantly increased ROS within 30 min (Koutsogiannaki et al. 2014). An EEDCs mixture, including E2, caused increased aldehyde (e.g., malondialdehyde, MDA) levels in the digestive gland of mussels 24 h after injection into the posterior adductor muscle (Canesi et al. 2008). Similarly, dietary exposure of MPs in mussels induced oxidative stress in the digestive gland, confirmed by the significant production of MDA in this tissue (Wang et al. 2021).

The oxidation of lipids by ROS results in the formation of carbon cantered lipid radicals (L·) and lipid peroxide radicals (LOO·), which then produce lipid hydroperoxides (LOOH). LOOH can be eliminated through production of antioxidant enzymes, but if not, LOOH can decompose into highly reactive lipid peroxyl (LOO·) and/or alkoxyl (LO·) radicals, initiating a chain reaction (Alves de Almeida et al. 2007) in which more lipids are oxidized to generate ketones, epoxides, and aldehydes as a result of oxidative stress (Alves de Almeida et al. 2007; Ayala et al. 2014). Also, during adaptation to oxidative stress, oysters increase antioxidants and amino acids to counteract the stress (Ayala et al. 2014). In our previous study, several antioxidant amino acids (i.e., building blocks of antioxidant proteins), such as proline, glutamine, and tyrosine significantly increased in MP and estrogen exposed groups compared to controls (Kumar et al. 2024). Finally, the upregulation of phospholipid production indicates compensatory effects to lipid peroxidation due to MPs and EEDCs exposure.

Dietary influence on stress response

When organisms are under stress, they try to mitigate the stress through adaptive responses. For example, molluscs may close the valves (i.e., the two shell halves of the mollusc) to reduce pollutant exposure and increase antioxidant activity to mitigate the effects of pollutants. There was no indication of valve closure observed in this study, which may relate to the very low concentrations of both MPs (66 particles/L) and EEDCs used failing to elicit this response in our oysters. In mussel studies, exposure to 2000 particles/L of PVC-MP did not cause valve closure (Stollberg et al. 2021). Similarly, with clams, MP exposure concentrations below 2000 particles/clam had no significant effect on ingestion rate (Jessica et al. 2023). For EEDCs, our data indicated that even at high E2 (200 ng/L) and NP (5000 ng/L) exposures, they did not significantly change their feeding rate (Supplementary Fig. S1).

It is possible that in our study, oysters maintained regular or slightly increased food intake to mitigate the minor oxidative stress from PE-MP and EEDC exposure. Previous studies indicate that molluscs tend to mitigate stress by adjusting energy metabolism and antioxidant proteins that primarily come from algae (Ayala et al. 2014; Mai et al. 2023; Kumar et al. 2024). A slight increase in total cholesterols, especially in the male digestive gland of exposed groups (1.10–1.45 times higher compared to controls) may therefore reflect slightly higher algae consumption to mitigate stress, since oysters have little to no capability for cholesterol synthesis or conversion from phytosterols (Teshima and Patterson 1981; Holden and Patterson 1991) and this sterol must be acquired by ingesting food (Idler and Wiseman 1972; Berenberg and Patterson 1981). In this study, oysters were fed three algal species comprising 50% Chaetoceros meulleri, 25% Tisochrysis lutea, and 25% Diacronema lutheri. These algae species indeed contain phytosterols, primarily the cholesterols as their major sterol (Tsitsa-Tzardis et al. 1993; Ahmed et al. 2015; Bustamam et al. 2021). Although evidence is limited, the literature also suggests that the presence of food may help to mediate the effects of micro and nano plastics (PS-MPs at 70 nm and 10 μm), as observed in the thick shell mussel, Mytilus coruscus (Wang et al. 2021). A low concentration of particles failed to significantly reduce feeding and clearance rate as well as absorption efficiency of organics from algae in mussel (Wang et al. 2021). We also hypothesise that exposure to low, environmentally relevant concentrations of MPs and EEDCs induced transient oxidative stress in oysters that was mediated by regular food intake or a slight increase in food intake.

Co-exposure effects are negligible

Although the abundance of cholesterols and aldehydes was significantly increased in PE-MP and EEDC exposed groups, and were much higher in the EEDC treatment, there was no consistent pattern of significant increased response for co-exposure treatments compared to the single treatments. Also, pre-incubation did not consistently differ from simultaneous co-exposure of PE-MPs with EEDCs for cholesterols and aldehydes induction.

The negligible co-exposure effect might relate to the relatively low exposure concentrations of both PE-MPs and EEDCs. Hydrophobic organic contaminants, like EEDCs have different partition coefficients for MPs, water and microalgae (Li et al. 2025). Therefore, it is hypothesised that in co-exposure, EEDCs partitioned not just on PE-MPs, but also in seawater and on algae. The feeding behaviour of oysters may have reduced the exposure of EEDCs that were adsorbed on MPs. Being a selective feeder, oysters might have expelled the EEDC adsorbed PE-MPs before ingestion. A recent review concluded that the compound effects of MPs and organic pollutants depends on the ingestion or internalization of MPs by organisms (Ding et al. 2022). Also, exposure of hydrophobic organic compounds (i.e., EEDCs) is higher from natural food (e.g., algae) than from MPs (Koelmans et al. 2016). Indeed, other studies confirm that in molluscs, co-exposure with MPs may result in lower impacts than exposure to chemical pollutants alone. Even in co-exposure, low MP concentrations alleviated the effects of caffeine and chlorpromazine (Yunko et al. 2025). Similar negligible vector roles of MPs to clams were reported for polycyclic aromatic hydrocarbons, where oxidative stress related enzymes indicated that mixed exposure was less toxic than individual exposures (Kılıç et al. 2023). The negligible co-exposure effect of PE-MPs and EEDCs in this study were therefore likely due to the low exposure concentration of both, and the selective feeding pattern of oysters. Additionally, acute exposure, short term incubation and, more importantly, the smooth spherical MPs might have resulted in a lack of a combined stress response in oysters. Spherical PE-MPs (size: 10–45 μm) were also found to be a poor vector of benzo(a)anthracene to the Manila clam (Kılıç et al. 2023). It is also evident that when the surface of spherical PE-MPs was changed by etching, it significantly enhanced sorption of polyaromatic hydrocarbons by to hydrogen bonding and π-π interactions (Li et al. 2020).

Limitations and future research directions

While this study provides novel lipidomic insights into the effects of environmentally relevant exposure to PE-MPs and EEDCs in oysters, several limitations should be acknowledged. First, the exposure duration was relatively short (7 d) and focused on acute to short-term responses. Although this timeframe was sufficient to capture transient oxidative stress and adaptive lipidomic responses, longer-term or chronic exposures may reveal cumulative effects, delayed toxicity, or shifts from adaptive to adverse outcomes, particularly under fluctuating environmental conditions.

Second, lipidomic analyses were conducted on the digestive gland of both sexes and the female gonad only. Male gonadal tissue was not included, as previous studies using the same experimental framework demonstrated limited responsiveness of male gonads to estrogenic mixtures. Nonetheless, inclusion of male gonad lipidomics in future studies would provide a more comprehensive understanding of sex-specific reproductive sensitivity, especially across different reproductive stages or under higher contaminant loads.

Third, although MP uptake was verified by visual examination of the gut, quantitative mass-balance assessments of MP ingestion, retention, and egestion were limited. As selective feeding and pseudofeces production likely played an important role in moderating exposure, future studies should quantify MP uptake rates, residence time, and egestion dynamics to better link internal MP burdens with biochemical effects and to refine assessments of MP vector potential.

Fourth, this study used smooth, spherical, commercially available PE-MPs, which may underestimate the vector role of environmentally aged or irregular MPs. Aging, surface roughness, and bio-corona formation are known to modify sorption–desorption dynamics of hydrophobic organic contaminants. Future research should therefore examine aged MPs, environmentally sourced particles, and different polymer types to better reflect real-world exposure scenarios.

Finally, lipidomic changes were interpreted primarily in the context of oxidative stress and dietary compensation. Integration of lipidomics with targeted measurements of antioxidant enzymes, lipid peroxidation biomarkers, feeding rates, and energy reserves would strengthen mechanistic interpretation. Multi-omics approaches combining lipidomics, metabolomics, and transcriptomics would be particularly valuable in resolving adaptive versus adverse responses under low-dose, mixed-contaminant exposure.

Conclusion

Environmental exposure to PE-MP and EEDC induced oxidative stress in oysters and a consequent increase in lipid peroxidation products in some tissues, but our results suggest that they may have largely adapted to mitigate the effects. This likely relates to the low, environmentally relevant exposure scenario, where MPs and EEDCs induce only mild sub-lethal toxicity in oysters. The resulting transient oxidative stress response would be easily mitigated by the oysters by maintaining regular food intake or slightly increasing intake to replenish oxidised/depleted lipid stores (as evidenced by an increase in phospholipids). Selective feeding behaviour may have also helped to alleviate MP intake, and consequent uptake of EEDCs adsorbed on MPs surface. While the exact explanation requires further confirmatory research, our results indicate that co-exposure effects of MPs and EEDCs at environmentally relevant levels are negligible. Pre-incubation of EEDCs with MPs did not facilitate the availability of EEDCs to oysters, with oysters probably rejecting MPs as pseudofeces before assimilating them into the digestive system. Together, these results suggest that MPs are not an efficient vehicle of EEDCs in molluscs at low concentrations of MPs and EEDCs, with continued supply of food, and due to the smooth surface of commercially available PE-MPs.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1