Nanoplastics increase in vitro oestrogenic activity of neurotherapeutic drugs

PSNPs were purchased as a commercial stock of 25 nm particles in the concentration of 10,500 mg/L from Phosphorex (Hopkinton, MA, USA). They were visualised and their primary size (d, nm) determined with a transmission electron microscope (TEM) (JEOL JEM 1010, JEOL, Tokyo, Japan) in samples prepared by suspending them in the cell culture RPMI-1640 medium without phenol red (Sigma Aldrich, Steinheim, Germany) supplemented with 5 % charcoal-stripped foetal bovine serum (CS-FBS) (Sigma Aldrich) to obtain the concentration of 1 mg/L. The suspension was dropped on a Formvar^®-coated copper grid (SPI Supplies, West Chester, PA, USA), dried overnight at room temperature, and measured in a bright field mode at an acceleration voltage of 80 kV. Images were taken with a Canon PowerShot S50 Camera (Canon, Tokyo, Japan). Results are presented as mean values (d, in nm) of 60 particles with standard deviations (SD) calculated with the ImageJ software (LOCI, University of Wisconsin, Madison, WI, USA). Hydrodynamic diameter (d_H) and size distribution were determined for PSNPs suspended in cell culture for 24 h and those not suspended in cell culture with dynamic light scattering and zeta potential using a Zetasizer Ultra instrument (Malvern Panalytical, Malvern, UK). Data were processed with the ZS Xplorer 3.21 software (Malvern Panalytical). The d_H results represent the means of six measurements expressed as intensity-weighed size distribution (in nm), while the zeta potential represents the means of three ELS measurements expressed in mV.

The T47D-KBluc cell line was purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). These are reporter-labelled cells transfected with triplet oestrogen-responsive elements used to screen for oestrogenic or anti-oestrogenic activity of chemicals (30). T47D-KBluc cells were cultured in tissue culture flasks (Sarstedt, Nümbrecht, Germany) with the RPMI-1640 medium supplemented with 10 % (v/v) CS-FBS and 1 % (v/v) antibiotic-antimycotic solution (Sigma Aldrich) at 37 °C and 5 % CO₂ to reach the density of 1×10⁶ cells/mL (90–95 % confluence) before treatment.

Before we determined oestrogen receptor activity, we established dose-response cytotoxicity for PSNPs, Cbz, and Flx alone and for their mixtures. For that purpose, T47D-KBluc cells were seeded in 12-well plates (Sarstedt, Nümbrecht, Germany) containing 1 mL of cell culture medium at a density of 1×10⁵ cells/well at 37 °C and 5 % CO₂. After 24 h, we replaced the medium and treated the cells with different concentrations of test substances or their mixtures.

Dose ranges were selected based on literature data on in vitro toxicity of PSNPs, Cbz, and Flx (31,32,33) as follows: 0.1–10 mg/L for PSNPs, 3.75–500 µmol/L for Cbz, and 0.75–100 µmol/L for Flx. Their cytotoxicity was first tested with the MTS assay as described elsewhere (34) with the aim to find the doses at which >90 % cells remained viable. Negative controls were untreated cells and positive controls were cells treated with 10 % (v/v) DMSO (Sigma Aldrich).

After 48 h of treatment at 37 °C and 5 % CO₂, the medium was removed from the plates into 2 mL Eppendorf tubes (Eppendorf, Hamburg, Germany). Cells remaining in the wells were washed with phosphate buffered saline (PBS) three times and then detached by incubating them with Trypsin-EDTA solution (Sigma Aldrich, Steinheim, Germany) at 37 °C and 5 % CO₂ for 5–7 min. The detached cells were then added to the Eppendorf tubes with previously collected medium.

Due to possible interferences with the MTS assay, cytotoxicity testing was then repeated using flow cytometry for PSNP doses of 0.1, 1, and 10 mg/L, for Cbz doses of 50 and 100 µmol/L, and for Flx doses of 5 and 10 µmol/L. Tube content (with viable and dead cells) was stained with Annexin V-FITC (AnnV) and propidium iodide (PI) using the flow cytometry Annexin V Kit (Bio Rad, Hercules, California, USA) to count live (AnnV−, PI−), early apoptotic (AnnV+, PI−), apoptotic (AnnV+, PI+), and dead cells (AnnV−, PI+) on a Cytoflex SRT sorter using its software (Beckman Coulter Life Sciences, Indianapolis, IN, USA).

Results are reported as the percentage of live, early apoptotic, late apoptotic, or dead cells compared to negative controls obtained from three independent experiments done in triplicate.

For further experiments we used only Cbz, Flx, and PNSP doses that left more than 90 % of cells viable to avoid bias that may arise from dead, unviable, or damaged cells.

Oestrogen receptor activity in T47D-KBluc cells was determined for the test substances or their mixtures using the luciferase assay (Promega, Madison, WI, USA, Cat. Nos.: E1500 and E1501). The assay is based on the binding to oestrogen-responsive elements by oestrogen receptors that make part of the T47D-KBluc cell DNA sequence. Binding to these elements induces gene transcription that produces luciferase, which converts beetle luciferin (assay reagent) to a luminescent product oxyluciferin. The intensity of its luminescence is proportional to oestrogen receptor activation.

Our measurements followed the OECD test No. 455 (5). The responsiveness of the test system was checked using diethylstilbestrol (DES) as positive control and fulvestrant as negative control. The quality control of the assay confirmed ≥four-fold mean luciferase activity compared to negative/vehicle control on each plate and no interference by the test substances (PSNPs and neurodrugs) with the assay components and readouts.

The assay was run on T47D-KBluc cells that were first cultured for a week in a medium in which 10 % FBS was replaced by 10 % charcoal-stripped FBS to attenuate interferences from serum hormones. After one week, the cells were seeded in white opaque flat-bottom Nunc™ MicroWellTM 96-well microplates (Thermo Fisher Scientific, Waltham, MA, USA) at a density of 2×10⁴ cells per well, each containing 100 µL of culture medium containing 5 % (v/v) charcoal-stripped FBS and kept there at 37 °C and 5 % CO₂ for 24 h to attach to the wells. Followed a 48-hour treatment with different concentrations of PSNPs, Cbz, and Flx alone or their mixtures. Untreated cells were used as negative control and cells treated with 10 nmol/L DES as positive control. After the treatment, oestrogen receptor activity was measured on a SpectraMax iD3 microplate reader (Molecular Devices, San Jose, CA, USA) using the Promega luciferase assay kit as described above.

Results are expressed either as the percentage of fold luminescent signal inductions compared to negative or positive controls.

Statistical analysis was run on GraphPad Prism6 (GraphPad Software, San Diego, CA, USA). Statistical significance was determined by one-way ANOVA followed by Dunnett’s multiple comparison test with negative control values and set to P<0.05.

Figure 1 shows the shape, size, size distribution, and zeta potential of the PSNPs used in this study. They were of spherical shape and primary diameter (d_TEM) of 25.6±3.2 nm, confirming the producer’s declaration (Phosphorex). Their hydrodynamic diameter (d_H) was greater than d_TEM due to hydration and protein corona shell formation on the nanosurface, while further increase in d_H after 48 h indicates slight aggregation due to increased ionic strength of the cell culture medium (RPMI-1640). The zeta potential was negative and also increased after 48 h.

Figure 1

The MTS assay showed that doses below 60 µmol/L for Cbz and below 12.5 µmol/L for Flx did not affect the viability of T47DKBluc cells (Figure 2), while the IC₅₀ values were 174.8 and 30.17 µmol/L, respectively.

Figure 2

Further flow cytometry showed that none of tested substances at given doses (0.1, 1, 5, and 10 mg/L for PSNPs; 1, 5, 10, 50, and 100 µmol/L for Cbz; and 0.1, 0.5, 1, 5, and 10 µmol/L for Flx) induced significant damage to the T47D-KBluc cells (Figure 3), so we proceeded with these doses in further luciferase assay experiments.

Figure 3

Endocrine disrupting effects of Cbz and Flx in mixtures with PSNPs were determined by comparing cell oestrogen receptor activity in respective mixtures (Cbz-PSNPs and Flx-PSNPs) with that of individual components (Cbz, Flx, or PSNPs). As described in the OECD test No. 455, a substance can be considered endocrine disruptor if oestrogen receptor response is equal to or exceeds 10 % of the response obtained with 10 nmol/L of DES. In view of these guidelines, the tested substances alone did not induce oestrogen receptor activity (Figure 4).

Figure 4

However, Cbz-PSNP and Flx-PSNP mixtures show quite alarming results in terms of health hazard and evidence that nanoplastics do behave like a Trojan horse in T47D-KBluc cells regardless of their concentration (Figure 5). Cbz and Flx mixed with PSNPs evoked strong endocrine disrupting response (≥10 % of oestrogen response to 10 nmol/L DES) starting with the doses of 50 and 1 µmol/L, respectively.

Figure 5

Interestingly, oestrogen receptor response did not depend on PSNP concentrations. This may be owed to greater PSNP aggregation in the cell culture medium when particle concentrations are higher, which may decrease specific surface area available for interaction with cells and drug absorption. Furthermore, the observed response did not show dose dependence for the two higher drug doses (50 and 100 µmol/L of Cbz, 5 and 10 µmol/L of Flx), most likely because of nanosurface saturation.