Microplastic exposure induces epithelial barrier alterations and increases collagen deposition in a 3D human endometrial model in vitro

HE staining demonstrates that hESCs infiltrated the scaffold inserts and acquired a 3D organization, generating a densely populated stromal compartment (Fig.  1 a). hEECs, seeded on top of it, formed a monolayer and became elongated or cuboidal in shape (Fig.  1 a). Furthermore, TEER measurements of the epithelial barrier integrity show increasing values, after hEEC seeding that peaked up to 157.63 ± 13.44 Ω*cm 2 on day 28 of culture and were stably maintained for the entire length of the experiments (Fig.  1 b). Fig. 1 Characterization of the 3D endometrial model. a Representative images of HE staining of the 3D endometrial model obtained by co-culturing hESCs and hEECs onto highly porous scaffolds (scale bars 100 and 25 μm). b TEER values detected at different time points. Data are expressed as the mean ± standard deviation (SD). a,b,c,d Different superscripts indicate p  < 0.05 Characterization of the 3D endometrial model. a Representative images of HE staining of the 3D endometrial model obtained by co-culturing hESCs and hEECs onto highly porous scaffolds (scale bars 100 and 25 μm). b TEER values detected at different time points. Data are expressed as the mean ± standard deviation (SD). a,b,c,d Different superscripts indicate p  < 0.05 Confocal microscopic analysis demonstrates the presence of fluorescent PS-MPs within the 3D endometrial models belonging to all the experimental groups (Fig.  2 a). In particular, PS-MPs crossed the epithelial barrier and were visible in the stromal compartment starting from 24-h exposure, regardless to the concentrations used. Fig. 2 PS-MP internalization and cytotoxic effects. a Three-dimensional z-stack sections of 3D endometrial models exposed for 24 and 48 h to different PS-MP concentrations. Nuclei are counterstained with DAPI (blue signal). Red dots indicate fluorescent PS-MPs. b MTT assay after 24-h exposure to different concentrations of PS-MPs. Data are expressed as the mean ± standard deviation (SD). c MTT assay after 48-h exposure to different concentrations of PS-MPs. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05 PS-MP internalization and cytotoxic effects. a Three-dimensional z-stack sections of 3D endometrial models exposed for 24 and 48 h to different PS-MP concentrations. Nuclei are counterstained with DAPI (blue signal). Red dots indicate fluorescent PS-MPs. b MTT assay after 24-h exposure to different concentrations of PS-MPs. Data are expressed as the mean ± standard deviation (SD). c MTT assay after 48-h exposure to different concentrations of PS-MPs. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05 However, MTT assay shows no cytotoxic effects exerted by PS-MPs after 24-h exposure (24 h, Fig.  2 b). In particular, no significant differences in OD values were detected between the 3D endometrial models incubated with 0.25, 0.5, 0.75, 1, 10, 12.5, 25 and 50 mg/ml PS-MPs and those of the control (CTR; Fig.  2 b). Similar results were observed for prolonged treatments (48 h) at lower PS-MP concentrations, ranging from 0.25 to 1 mg/ml (48 h, Fig.  2 c). In contrast, protracted exposures at higher concentrations—from 10 to 50 mg/ml—induced significant differences in OD values compared to the CTR (48 h, Fig.  2 c). HE staining demonstrates the maintainance of a robust stromal compartment and the persistence of an intact epithelial layer, after 24-h exposure to PS-MPs, regardless to the concentrations used (24 h, Fig.  3 ). Similarly, PS-MP prolonged exposure (48 h) at low concentrations—ranging from 0.25 to 1 mg/ml—did not induce any evident morphological alteration (48 h, Fig.  3 ). In contrast, higher PS-MP concentrations—from 10 to 50 mg/ml—applied for 48 h, induced visible damages to the epithelial layer that became thinner and discontinuous (48 h, Fig.  3 ). Fig. 3 Histological evaluation of the 3D endometrial models after PS-MP exposure. HE staining of the 3D endometrial models exposed for 24 and 48 h to different concentrations of PS-MPs (scale bars 100 and 10 μm) Histological evaluation of the 3D endometrial models after PS-MP exposure. HE staining of the 3D endometrial models exposed for 24 and 48 h to different concentrations of PS-MPs (scale bars 100 and 10 μm) Consistent with the morphological observations, TEER values are statistically comparable among the untreated groups (CTR) and all the endometrial models exposed to PS-MPs for 24 h (Fig.  4 a). In addition, TEER measurements detected in the control groups are also comparable to those of the 3D endometrial models incubated with 0.25, 0.5, 0.75 and 1 mg/ml of PS-MPs for 48 h (Fig.  4 b). Conversely, TEER values significantly decreased in the 3D endometrial models incubated for 48 h with 10, 12.5, 25 and 50 mg/ml of PS-MPs (Fig.  4 b). In agreement with this, transcription levels of ZO1 and CDH1 genes are statistically comparable after 24-h incubation, regardless to the concentrations used (Fig.  4 c), as well as when exposure was prolonged to 48 h with PS-MP concentrations ranging from 0.25 to 1 mg/ml (Fig.  4 d). Higher doses of PS-MPs, in contrast, resulted in a significantly decreased expression of both genes (Fig.  4 d). Confocal analysis confirms the decrease of ZO1 immuno-positivity, which, in turn, was clearly detectable in the epithelial layer of the untreated models (CTR, Fig.  4 e). Fig. 4 Functional evaluation of the 3D endometrial models after PS-MP exposure. a TEER values after 24-h exposure to different concentrations of PS-MPs. Data are expressed as the mean ± standard deviation (SD). b TEER values after 48-h exposure to different concentrations of PS-MPs. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05. c Gene expression analysis of ZO1 and CDH1 after 24-h exposure to different concentrations of PS-MPs. Gene expression is presented with the highest-level set to 1 and all others relative to this. Data are expressed as the mean ± standard deviation (SD). d Gene expression analysis of ZO1 and CDH1 after 48-h exposure to different concentrations of PS-MPs. Gene expression is presented with the highest-level set to 1 and all others relative to this. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05. e Representative confocal images of the untreated 3D endometrial models (CTR) and after exposure to different PS-MP concentrations for 48 h. Samples were stained for ZO1 (green) and DAPI (blue). PS-MPs are detected in yellow Functional evaluation of the 3D endometrial models after PS-MP exposure. a TEER values after 24-h exposure to different concentrations of PS-MPs. Data are expressed as the mean ± standard deviation (SD). b TEER values after 48-h exposure to different concentrations of PS-MPs. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05. c Gene expression analysis of ZO1 and CDH1 after 24-h exposure to different concentrations of PS-MPs. Gene expression is presented with the highest-level set to 1 and all others relative to this. Data are expressed as the mean ± standard deviation (SD). d Gene expression analysis of ZO1 and CDH1 after 48-h exposure to different concentrations of PS-MPs. Gene expression is presented with the highest-level set to 1 and all others relative to this. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05. e Representative confocal images of the untreated 3D endometrial models (CTR) and after exposure to different PS-MP concentrations for 48 h. Samples were stained for ZO1 (green) and DAPI (blue). PS-MPs are detected in yellow Picrosirius red staining indicates no visible variations in collagen content among the untreated samples (CTR), those exposure to PS-MPs for 24 h, regardless to the concentrations used, and those incubated for 48 h with concentrations up to 1 mg/ml of PS-MPs (Fig.  5 a). In contrast, samples exposed for 48 h to PS-MP doses ranging from 10 to 50 mg/ml show strong intensity of the staining (Fig.  5 a). This is also confirmed by stereological analyses that demonstrate a statistically significant increment in collagen content only in samples exposed for 48 h to PS-MP concentrations ranging from 10 to 50 mg/ml (Fig.  5 b). Gene expression analysis indicated a statistically significant upregulation of the transcription levels of the COL1A2, COL3A1 and COL4A2 genes only in the 3D endometrial models treated with 10, 12.5, 25 and 50 mg/ml PS-MPs for 48 h (Fig.  6 b). This is paralleled by a significantly decreased expression of the MMP1, MMP2 and MMP9 genes in the same sample (Fig.  6 b). Increased collagen transcription was accompanied by a significant upregulation of the TP53, WNT1, TGFB1 and SMAD3 genes (Fig.  6 b). Fig. 5 Histochemical and stereological analyses for collagen in the 3D endometrial models after PS-MP exposure. a Picrosirius red staining after 24- and 48-h exposure to different concentrations of PS-MPs (scale bars 100 µm); b Stereological analysis of collagen content. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05 Fig. 6 Molecular analysis for COL1A2, COL3A1, COL4A2, MMP1, MMP2, MMP9, TP53, WNT1, TGFB1 and SMAD3 genes in the 3D endometrial models after 24- ( a ) and 48-h ( b ) exposure to different concentrations of PS-MPs. Gene expression is presented with the highest-level set to 1 and all others relative to this. Yellow histograms show the cycle threshold (Ct) values of ACTB and GAPDH in all the experimental groups. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05 Histochemical and stereological analyses for collagen in the 3D endometrial models after PS-MP exposure. a Picrosirius red staining after 24- and 48-h exposure to different concentrations of PS-MPs (scale bars 100 µm); b Stereological analysis of collagen content. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05 Molecular analysis for COL1A2, COL3A1, COL4A2, MMP1, MMP2, MMP9, TP53, WNT1, TGFB1 and SMAD3 genes in the 3D endometrial models after 24- ( a ) and 48-h ( b ) exposure to different concentrations of PS-MPs. Gene expression is presented with the highest-level set to 1 and all others relative to this. Yellow histograms show the cycle threshold (Ct) values of ACTB and GAPDH in all the experimental groups. Data are expressed as the mean ± standard deviation (SD). a,b Different superscripts indicate p  < 0.05

All reagents were purchased from Thermo Fisher Scientific (Milan, Italy) unless otherwise indicated. Human endometrial stromal and epithelial cells (ABI-TC257D and ABC-TC4601) were purchased from Accugen (USA). This study did not involve the use of living humans, and therefore, ethical approval was not required. All the methods were carried out following the approved guidelines. 3D endometrial models were generated following a protocol previously described [ 51 ]. Briefly, endometrial stromal cells (hESCs) were triple seeded onto Alvetex® Scaffold inserts (Reprocell Europe, UK) and cultured in Dulbecco’s modified eagle medium (DMEM) supplemented with 10% bovine foetal serum (FBS), 2 mM glutamine, 5-ng/mL transforming growth factor beta (TGFβ), 100-μg/mL ascorbic acid and 1% penicillin–streptomycin-amphotericin B solution, for 14 days in 5% CO 2 at 37 °C. On day 14, 2 × 10 6 endometrial epithelial cells (hEECs) were seeded on the top of the stromal compartment and cultured in Dulbecco’s modified eagle medium/nutrient mixture F-12 (DMEM/F12) supplemented with 5% FBS, 2 mM glutamine and 1% penicillin–streptomycin-amphotericin B solution. Cultures were maintained for 21 days in 5% CO 2 at 37 °C. Medium was refreshed twice a week. All experiments were carried out with hESCs and hEECs at passages between 7–11 and 10–14, respectively. The generated 3D endometrial models were incubated for 24 and 48 h with 0.25, 0.5, 0.75, 1, 10, 12.5, 25 and 50 mg/ml Latex beads, amine-modified polystyrene, fluorescent red (PS-MPs, #L2778, Sigma-Aldrich) diluted in DMEM/F12 medium supplemented with 5% FBS, 2 mM glutamine and 1% penicillin–streptomycin-amphotericin B solution. These concentrations were selected based on preliminary experiments where 1 ug/ml, 5 ug/ml, 10 ug/ml, 20 ug/ml, 50 ug/ml and 100 ug/ml were tested with no response. Culture medium without PS-MPs was used as control (CTR). All exposures were performed at least in triplicate in three independent experiments. 3(4,5-Dimethylthiazole-2-yl)−2,5-diphenyltetrazolium-bromide (MTT, Roche) assay was performed on 3D endometrial models exposed to 0, 0.25, 0.5, 0.75, 1, 10, 12.5, 25 and 50 mg/ml PS-MPs for 24 and 48 h. Briefly, scaffold inserts were extensively washed in 1 × phosphate-buffered saline (PBS; Sigma-Aldrich) and transferred in new 12-well plates. One millilitre of MTT solution was then added to each well and incubated for 1 h. Formazan salt crystal products were dissolved in 10 ml of 10% SDS in 0.01 M HCl overnight. The optical density (OD) was measured at 550 nm. Wells without scaffold inserts were used as negative controls. Samples were fixed in 4% paraformaldehyde (Sigma-Aldrich) for 24 h, dehydrated through a series of ethanol, incubated in Histoclear (Bio-optica) and embedded in paraffin. Sections of 5–7-μm-thick were cut, dewaxed, re-hydrated and stained with haematoxylin/eosin (HE, Histoline) and Picrosirius red (Sigma-Aldrich), following the manufacturer’s instructions. Stained sections were analyzed under Leica DMR microscope (Leica Microsystems). Volume density (Vv) evaluations were performed on 30 sections obtained from each experimental group (ten sections for each replicate) and stained with Picrosirius red. The Delesse principle was used, and the proportional volume of each specific area was calculated as the fraction of the structure of interest relative to the total area of the reference compartment (e.g., whole section). Images were randomly taken, overlaid with a point-count stereological grid containing evenly spaced test points and the relative volume of each region of interest was calculated by dividing the number of points striking the structure of interest by the number of points hitting the reference compartment. Vv was expressed as percentages using the following formula: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text{Vv}\left(\text{analyzed compartment},\text{ reference compartment}\right)=\left[\sum P\left(\text{analyzed compartment}\right)/\sum P\left(\text{reference compartment}\right)\right]\times 100$$\end{document} Vv analyzed compartment , reference compartment = ∑ P analyzed compartment / ∑ P reference compartment × 100 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sum P$$\end{document} ∑ P (analyzed compartment): the number of points hitting the compartment under study; \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sum P$$\end{document} ∑ P (reference compartment): the number of points hitting the relevant structure. EVOM2 Epithelial Voltmeter with STX3 electrode (World Precision Instrument) was used to measure TEER on the 3D endometrial models. Resistance was measured at three equidistant points of the insert perimeter, and the average was calculated. A blank insert, without cells, was run as a control, and the final TEER value was determined as follows: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text{TEER}\left(\text{Ohm}\times {\text{cm}}^{2}\right)=\left(\text{TEER average sample insert}-\text{TEER average blank insert}\right)\times \text{area }{\text{cm}}^{2}$$\end{document} TEER Ohm × cm 2 = TEER average sample insert - TEER average blank insert × area cm 2 Samples were fixed in 4% paraformaldehyde for 20 min, washed three times in 1 × PBS (Sigma-Aldrich), and permeabilized with 0.3% of Triton-X100 (Sigma-Aldrich) for 30 min. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Samples were visualized using a FRET-FLIM confocal microscope (Nikon) equipped with a digital camera (Nikon). Samples were fixed in 4% paraformaldehyde for 20 min, washed three times in 1 × PBS (Sigma-Aldrich) and treated with a blocking solution containing 5% bovine serum albumin (BSA; Sigma-Aldrich) and 0.3% of Triton-X100 (Sigma-Aldrich) in PBS for 30 min. Overnight incubations with 1:100 ZO-1 monoclonal antibody Alexa Fluor™ 488 conjugated (#339188) diluted in blocking solution were carried out at + 4 °C. The following day, nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Samples were visualized using a FRET-FLIM confocal microscope (Nikon) equipped with a digital camera (Nikon). TaqManGene Expression Cells-to-CT kit was used to extract RNA from the samples, according to the manufacturer’s instruction. DNase I (1:100 dilution) was added in lysis solution. Quantitative real-time PCR was performed on reverse transcribed samples using predesigned gene-specific primers and probe sets from TaqManGene expression assays (details are reported in Table  1 ). CFX96 real-time PCR detection system (Bio-Rad Laboratories) and CFX Manager software (Bio-Rad Laboratories) were used to quantify target genes. GAPDH and ACTB were used as internal standard references. Gene transcription levels are reported with the highest expression set to 1 and the other relative to this. Table 1 List of Taqman probes used for quantitative PCR Gene Description CAT. N ACTB Actin β Hs01060665_g1 CDH1 Cadherin 1 Hs01023895_m1 COL1A2 Collagen type I alpha 2 chain Hs01028956_m1 COL3A1 Collagen type III alpha 1 chain Hs00943809_m1 COL4A2 Collagen type IV alpha 2 Hs05006309_m1 GAPDH Glyceraldehyde-3-phosphate dehydrogenase Hs02786624_g1 MMP1 Matrix metallopeptidase 1 Hs00899658_m1 MMP2 Matrix metallopeptidase 2 Hs01548727_m1 MMP9 Matrix metallopeptidase 9 Hs00957562_m1 TP53 Tumour protein p53 Hs01034249_m1 SMAD3 SMAD family member 3 Hs00969210_m1 TGFB1 Transforming growth factor beta 1 Hs00998133_m1 List of Taqman probes used for quantitative PCR Statistical analysis was performed using two-way ANOVA test (SPSS 19.1; IBM). Data are presented as mean ± standard deviation (SD). Differences of p  ≤ 0.05 were considered significant and different superscript letters in the graph and histograms indicate significant differences between groups ( p  ≤ 0.05).

In the present study, we demonstrate MP ability to exert deleterious effects on a human endometrial model in vitro and, specifically, to cause epithelial barrier integrity alterations and to induce the activation of pro-fibrotic signaling pathways, resulting in collagen synthesis and deposition increment. The 3D human endometrial barriers, described here, are generated following a protocol recently described, based on which, hEECs and hESCs are co-cultured onto 3D highly porous scaffolds [ 51 ]. At the end of the culture period, histological analyses reveal the formation of a densely populated stromal compartment and the presence of a monolayer, on top of it, with epithelial cells that lose the flat morphology distinctive of 2D systems, and acquire an elongated or cuboidal shape, typical of differentiated polarized cells in vivo [ 52 ]. This is consistent with previous studies which reported the generation of in vitro artificial endometrial models in species other than the human, such as the bovine [ 53 ] and in the porcine [ 51 ] and that were shown to faithfully recreate the topography of the original tissue. In addition, the successful generation of a functional organotypic epithelial barrier, with good integrity, is confirmed by TEER measurements that display an average value of 157.63 ± 13.44 Ω*cm 2 from day 28 of culture onward. Unfortunately, to our knowledge, no data on the endometrium electrical resistance in vivo are currently available in the literature. However, these values are consistent with previous in vitro studies carried out using human and monkey oviductal epithelial cells which showed similar TEER measurements, when cultured onto 3D scaffolds [ 54 , 55 ]. These data indicate that the generated 3D endometrial model is able to mimic in vitro the complex architecture and physiology of its native counterpart in vivo and represents a useful bio-engineered tool to better elucidate the complex mechanisms taking place in the reproductive tract. Although this platform can provide more physiological relevant information than those obtained with 2D cultures, its complexity can be a challenge when scaling up for large scale studies are needed. In addition, imaging can be technically demanding. However, when the required expertise is achieved, highly informative data can be obtained. In the present study, confocal microscopic images display the presence of fluorescent PS-MPs within the 3D endometrial models starting from 24-h exposure, regardless to the concentrations used. In particular, PS-MPs were visible in the stromal compartment, suggesting that they are internalized by the epithelial barrier. However, MTT assay demonstrated that only protracted exposures at higher MP concentrations—from 10 to 50 mg/ml—induced significant differences in cell viability and proliferation, while no differences were detected among the other experimental groups and the untreated models. In agreement with this, exposure of the generated 3D endometrial models to MPs for 24 h causes no evident alterations on the endometrial barrier integrity, regardless of the concentration tested. This is demonstrated by cytotoxicity tests, histological staining as well as by TEER measurements, which show no significant differences in OD values between 3D models exposed to PS-MPs and those of the untreated group (CTR), the maintenance of a robust stromal compartment, the persistence of an intact epithelial layer and TEER values comparable to the control group (CTR). Our results are consistent with Lehner et al. and Chen et al. which described no cytotoxic/inflammatory effects nor barrier integrity damages caused by MP exposure of 3D human intestine models [ 56 , 57 ]. Similarly, no TEER changes were observed when advanced epithelial lung and gut barrier models were exposed to different concentrations, sizes and types of MPs for 24 h [ 58 ]. In addition, the molecular results presented in our manuscript support the morphological and functional observations, showing the maintenance of comparable transcription levels of the ZO1 and CDH1 genes, among all the 24-h treated groups and the control ones. This is in agreement with a previous study reporting no alterations in the expression levels of the tight junction related genes ZO1, OCLN and CLDN1, in a Caco-2 monolayer model [ 59 ]. Similar responses are detected, in our experiments, when concentrations between 0.25 and 1 mg/ml are used for 48 h. This is in line with Li et al. that tested similar concentrations and time of exposure in human retinal epithelial cells and reported no significant changes in cell viability [ 60 ]. In contrast, in the present study, 48-h incubations with higher doses, ranging from 10 to 50 mg/ml, induce cytotoxic effects, epithelial barrier alterations, reduce TEER values and decrease ZO1 and CDH1 gene transcriptions. This suggests that MPs in this size, concentrations and time of exposure exert harmful effects in vitro on the human endometrial barrier. It must be noted that selecting MP concentrations to be used in in vitro experiments is not an easy task, due to the lack of accurate human exposure data [ 61 ] that makes difficult the identification of “standard/physiological” concentrations. In particular, diverse effects of MP toxicity have been reported, depending on the specific cell type, which have been shown to possess a different vulnerability [ 62 ]. This variability is further amplified by the differentiation state of the cells as well as by the complexity of the in vitro model used. In addition, it is essential to note that, due to the limited knowledge about endometrial exposure concentrations in vivo, it is not possible to hypothesize how realistic our exposure conditions are. It is, however, of interest that, beside the effect on the epithelial layer, 48-h exposure with doses higher than 1 mg/ml, also affected the stromal compartment, inducing the activation of pro-fibrotic signalling pathways that results in collagen synthesis and deposition increment. Clear evidence of fibrosis was previously observed in uteri of mice exposed to MPs via drinking water [ 41 ] as well as in rats that displayed fibrotic ovaries after MP administration [ 63 ] . Altogether these data are in line with the pro-fibrotic effect detected in the present study and call attention for the potential risks and subsequent adverse effects on the human reproductive system, due to environmental exposure to MPs.

Despite a number of evidence suggests that MPs may significantly impact fertility, the adverse effects activated in the female reproductive system need to be further elucidated, with both in vivo as well as through the use of predictive in vitro models. Altogether, the data obtained in the present study indicate MP ability to exert deleterious effects in vitro on human endometrium, with a negative impact on the epithelial barrier integrity that may alter its functionality and receptivity. Therefore, although plastic offers considerable benefits, efforts are mandatory to reduce plastic use, to improve waste management in an environmentally sound manner, and to implement novel preventive strategies to minimize the impact of MP exposure on fertility.

Plastic pollution is a steadily increasing global phenomenon that promotes climate and environmental changes, negatively affects food safety, thus impairing human and animal health [ 1 – 3 ]. In the last few decades, thanks to its properties, such as lightweight, high strength, versatile fabrication, design and corrosion resistance [ 4 , 5 ], plastic became one of the most widely used materials in different industrial activities [ 6 ]. On the other hand, plastic is extremely difficult to degrade naturally, and many countries have neither advanced technological facility or proper rules and regulations for waste management [ 7 ]. This results in a rapid increase of plastic accumulation in various natural habitats, including oceans, air and soil [ 8 – 14 ]. Aggravating this issue is the ability of environmental factors, like sunlight, heat, chemical/biological activities, to physical fragment plastic in particles of different sizes, which are known as macro (> 25 mm), meso (5–25 mm), large micro (1–5 mm), small micro (20 μm–1 mm) and nano (1–1000 nm) plastics [ 15 – 17 ]. Among them, large and small microplastic (MP) [ 18 ] as well as nanoplastic (NP) [ 19 , 20 ] pollution represents a substantial and multifaceted risk for ecosystems and human well-being [ 18 ]. Indeed, due to the small size, both MPs and NPs can easily enter the organisms through inhalation, ingestion or skin contact [ 1 , 21 – 27 ]. Once in the body, they can rapidly migrate through the bloodstream, thanks to their hydrophobic nature, and reach lymph nodes, other tissues and organs, causing different disorders, including cardiovascular and chronic kidney disease, birth defects and cancer [ 28 – 34 ]. Recent studies have demonstrated that MPs interfere also with the reproductive system, altering the blood-testis barrier, impairing spermatogenesis in males and causing placental dysfunctions, ovarian atrophy, endometrial hyperplasia and fibrosis in females [ 35 ]. In particular, Amran et al. reported that MP exposure is correlated with a decreased uterine weight, thinner endometrial layers, lower number of endometrial glands and reduced endometrial epithelial cell viability, in the rat [ 36 ]. Moreover, MP ability to induce murine endometrial adhesions, extensive cavitation as well as uterine cavity narrowing and fibrosis was reported [ 37 – 42 ]. Parallel studies also indicated that NP long-term exposure is associated with reproductive dysfunctions [ 24 , 40 – 43 ] and reduced reproductive performances [ 19 , 20 ]. For instance, various studies showed that NP significantly reduce reproductive-related hormone levels [ 44 – 46 ] and induce sperm cell apoptosis [ 47 ] as well as follicle atresia and uterus and endometrium alterations [ 48 ]. Although preliminary observations indicated the presence of MPs in endometrium and myometrium [ 49 , 50 ], only scattered information are available in the human. Considering the great importance of better understanding MP effects on the female reproductive system, it is essential to create appropriate high predictive in vitro models to study MP impact on the human endometrium and its physiology. To this purpose, in the present manuscript, we generated an in vitro 3D endometrial platform, co-culturing human endometrial stromal and epithelial cells onto highly porous scaffolds. The obtained 3D endometrial models were then exposed to different concentrations of MPs for 24 and 48 h. Histological staining, trans-epithelial resistance (TEER) measurements and gene expression studies were carried out to investigate MP effects on cell morphology, epithelial barrier integrity and collagen synthesis and deposition processes.