Internalized polystyrene nanoplastics trigger testicular damage and promote ferroptosis via CISD1 downregulation in mouse spermatocyte

Abstract

Background

There is a growing body of research regarding the potential reproductive toxicity of microplastics and nanoplastics. However, the underlying mechanisms by which nanoplastics exposure adversely affects the testes remain poorly understood. Our study aims to clarify the relationship between ferritinophagy and mitochondrial dysfunction based on polystyrene nanoplastics (PS-NPs)-caused testicular damage in mice.

Results

The current study demonstrates that 50 nm PS-NPs accumulate in mouse testes and lead to a decrease in sperm quality and disruption of spermatocyte. Furthermore, PS-NPs trigger ferroptosis in GC-2 cells, which can be mitigated by deferiprone and 3-methyladenine. Further investigation reveals that PS-NPs initially aggregate in lysosomes and subsequently transfer to the mitochondria. This process increases mitochondrial Fe²⁺ and mitochondrial ROS levels, as well as reduces the expression of CISD1, a protein that inhibits the uptake and transport of Fe²⁺ into the mitochondrial matrix. These changes ultimately result in disturbances to mitochondrial structure and function. In terms of mechanism, pioglitazone, a drug that stabilizes CISD1, has been demonstrated to mitigate ferroptosis induced by NCOA4-mediated ferritinophagy in GC-2 cells.

Conclusions

Our results indicate that PS-NPs cause mouse testicular damage through ferroptosis. Mechanistically, we confirmed that PS-NPs trigger NCOA4-mediated ferritinophagy and CISD1 downregulation in spermatocyte, which aggravates the flow of ferrous iron from the cytoplasm to the mitochondria.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03620-7.

Introduction

Plastic products are extensively utilized in various fields, including food, feed, and medicine, due to their exceptional durability and low cost [1]. However, the inadequate recycling efforts and absence of regulatory frameworks for the management of plastic waste have led to the pervasive distribution of plastics in terrestrial, marine, and atmospheric ecosystems globally. Over time, these plastic fragments undergo weathering and degradation processes, resulting in the formation of microplastics (MPs) and nanoplastics (NPs) [2]. These plastics have potential to infiltrate humans and animals through inhalation, ingestion, and skin exposure, resulting in harm to various tissues and organs. The discovery of MPs and NPs in placenta, blood, and follicular fluid has raised great concerns regarding their potential risks to human and animal health [3–5]. Of particular concern is the reproductive system, which is especially vulnerable to the effects of these particles [6, 7]. In comparison to MPs, NPs are distinguished by their larger surface area, smaller volume, and higher biological permeability, which could have the potential to pose a higher biological risk [8]. The presence of pollution in MPs and NPs has been regarded as a critical factor in global decline in fertility rates [9]. Moreover, there are many the additives in plastic including bisphenols and phthalates, identified as endocrine disruptors, NPs may act as the similar role to disrupt the reproductive hormones. A previous study identified the presence of MPs in testicular and seminal samples, with polystyrene MPs being the predominant type found in human testes [10]. Previous studies have demonstrated the risks for MPs and NPs to cause damage to spermatogenic cells, disrupt the blood-testis barrier, and diminish male fertility [11–13]. However, there remains a great gap in understanding of the toxic mechanisms by which polystyrene NPs (PS-NPs) affect male reproductive function.

As is well known, iron plays a vital role in the processes of spermatogenesis and testosterone synthesis [14]. There is a close relationship between iron homeostasis and testicular damage [14]. Ferroptosis, an iron-dependent programmed cell death type, was characterized by iron overload and lipid peroxidation [15]. Accumulating studies have indicated that ferroptosis is regulated by various signals, involving deactivation of Solute carrier family 7 member 11 (SLC7A11), glutathione (GSH) deficiency, cellular redox state, iron overload, acyl-CoA synthetase long-chain family member 4 (ACSL4)-mediated membrane PUFA-PL peroxidation and deactivation of glutathione peroxidase 4 (GPX4) [16]. Among these mechanisms, ferritinophagy plays an essential role in sustaining iron homeostasis and regulating ferroptosis [17]. Through the utilization of Nuclear Receptor Coactivator 4 (NCOA4) as a specific transport receptor, Ferritin Heavy Chain 1 (FTH1) is identified by autophagosomes and transported to the lysosome for degradation, resulting in the upregulation of intracellular iron content and the induction of oxidative stress, ultimately leading to ferroptosis [18]. Ferroptosis is associated with alterations in mitochondrial structure and function, with impaired mitochondria further exacerbating cellular susceptibility to ferroptosis [1]. CDGSH Iron Sulfur Domain 1 (CISD1), a mitochondrial outer membrane protein, takes part in the transportation of mitochondrial Fe²⁺. CISD1 is involved in hearing loss, ischemic reperfusion-injury, breast cancer, and other diseases [18–20]. Reduction of CISD1 expression detrimentally influences ferroptosis by giving rise to mitochondrial lipid peroxidation [22]. Although previous research has highlighted the crucial role of CISD1 in maintaining mitochondrial homeostasis, its relationship with ferritinophagy has yet to be investigated.

Previous studies have shown that a strong link between exposure of PS-MPs and ferroptosis in many organs. Li et al. [23] found Hypoxia-inducible factor-1 (HIF-1) and ferroptosis participated in PS-NPs-induced cardiotoxicity, involving ferrous iron accumulation, GPX4 downregulation, and ACSL4 upregulation. In addition, PS-NP exposure caused ferroptosis in bronchial epithelial cells by activating the HIF-1α/HO-1 signaling pathway, and eventually led to lung injury [24]. The study of Liang et al. [25] indicated that PS-NPs induced ferroptosis in Nrf2-deficient intestinal epithelial cells through ether phospholipid accumulation. For kidney, PS-NPs enhanced the sensitivity of cells to ferroptosis by adsorbing transferrin and facilitating its cellular internalization [26]. An in vitro study demonstrated that PS-NPs exposure induced ferroptosis via oxidative stress-triggered lipid peroxidation in oocytes, suggesting PS-NPs caused female reproductive toxicity [27], however, whether PS-NPs cause male reproductive system injury remains to be clarified.

The current study aims to clarify the relationship between ferritinophagy and mitochondrial dysfunction based on PS-NPs caused testicular damage through in vivo and in vitro experiments. Fluorescently labelled PS-NPs were employed to provide direct evidence of their dynamic distribution and accumulation in spermatocyte. To comprehensively determine the ferroptosis, we quantified intracellular ferrous ion, lipid peroxidation level and expression of ferroptosis-related proteins. Additionally, overexpression of NCOA4 and intervention with pioglitazone, a drug that increases the stability of CISD1, were used to investigate the impact of CISD1 downregulation and ferritinophagy on PS-NPs-triggered ferroptosis in spermatocyte. This study offers novel insights into the mechanism of PS-NPs-induced male reproductive toxicity and highlights the need for strategies to mitigate these reproductive risks in animals.

Materials and methods

Animals

Twenty specific pathogen-free (SPF) male KM mice were purchased from Chengdu Dossy Experimental Animals Co., LTD (Chengdu, China). The mice, 5 weeks old, were housed in the Northwest A & F University Laboratory Animal Center with pathogen-free conditions at 25 ± 1 °C with humidity of 55 ± 10%, and a 12 h light/dark cycle. Standard diet and germ-free water were freely available.

Polystyrene nanoplastics (PS-NPs, 50 nm) were obtained from Zhichuan Technology Co., Ltd. (Changzhou, China). Before oral administration, PS-NPs were suspended in ddH₂O and sonicated for 5 min (a cycle including sonication for 3 s and interval for 5 s) using Ultrasonic Cell Crusher (JY98-IIIN, Ningbo, China) in an ice bath. In order to explore the impact of PS-NPs on male reproductive system, twenty mice were randomized into 2 groups (10 mice/group). After 7-day adaptation, mice were exposed to 50 mg/kg PS-NPs in 200 µL solution by oral gavage, and mice of the negative control group were administrated orally with ddH₂O daily. The gavage lasted for 35 days until the mice were euthanized. Blood was collected behind the eyes, and placed at 4 °C for 2 h and centrifugated at 4000 rpm for 10 min at 4 °C to obtain serum. After decapitation, mouse testes and epididymides were collected for subsequent experiments.

Dose selection

Given that polystyrene is one of the three most prevalent polymers in environmental plastic particles, PS-NPs were employed in this study [28]. Additionally, PS is the predominant polymer found in the testes [10]. It is estimated that when body weight of a human is approximately 60 kg, the daily intake of plastic particles ranges from 2.4 to 700 mg [29]. The dose-by-factor method was employed to conduct dose conversion. Based on body surface area, the conversion formula was: Mouse equivalent dose (mg/kg) = Human dose (mg/kg) × [Km(human) / Km(mouse)]. The correction factor (Km) = Weight (kg) / body surface area (m²) [30]. The average weight of the human body is approximately 60 kg, compared to the average initial body weight of the mice used in this study is around 0.04 kg. The corresponding body surface areas are 1.62 m² for humans and 0.011 m² for mice. Consequently, the Km values for humans and mice are approximately 37.0 kg/m² and 3.6 kg/m², respectively [30]. With the estimated daily intake ranges from 0.04 to 11.7 mg/kg for humans (~ 60 kg), the equivalent dose for mice is translated to around 0.41 to 120.25 mg/kg. In this study, a dose of 50 mg/kg was chosen according to the previous studies [23, 24], which lies within the range of environmentally relevant concentrations [31].

Characterization of PS-NPs

Transmission electron microscopy (TEM) TECNAI G2 SPIRIT BIO (FEI, Hillsboro, USA) was utilized for the characterization of PS-NPs. The hydrodynamic diameter and zeta potential were measured by using a dynamic light scattering (DLS) was employed on a Zetasizer ZEN3600 (Malvern, Worcestershire, U.K.).

Sperm quality parameters analysis

Sperm quality parameters analysis was performed as described before with minor modifications [32]. The cauda epididymis of mouse was shredded and placed in 500 µL of preheated PBS at 37 °C, Subsequently, mouse sperm were carefully collected from the suspension and incubated in a water bath at 37 °C for 1 min. Pre-warmed disposable counting chamber slides (Minitube, Tiefenbach, Germany) were used to evaluate the sperm density, viability, motility, and abnormality rate with 3 µL of each sperm sample by Computer Assisted Sperm Analysis (CASA) system (Minitube, Tiefenbach, Germany).

Histological analysis of testis

4% paraformaldehyde was used to fix the isolated testes at room temperature overnight, and these testes were embedded in paraffin. Sections of the testes were then cut to a thickness of 2 μm, deparaffinized, and rehydrated. The slides were immersed in hematoxylin and shaken for 2 min, followed by 1-min rinse in H₂O. Subsequently, the slides were stained with 1% eosin solution, agitated for 3 s, and subsequently dehydrated in 95% and 100% alcohol twice for 3 min, respectively. Alcohol was extracted by two xylene washes. Finally, neutral resin was used to embed into the sections. Observation and image capture of the sections were performed using a BX40 microscope (Olympus, Tokyo, Japan). The histopathological evaluation, employing the Johnsen score system [33], was methodically carried out across ten distinct fields within each section.

Py-GC/MS analysis of PS-NPs content

Based on previous methods [10], the PS-NPs content in the mouse testes was evaluated using Py-GC/MS (Shimadzu, Kyoto, Japan). Briefly, the testicular tissue was digested with acid, and the treated solution was concentrated and dropped into the sampling crucible for analysis. The Py-GC/MS calibration curve for the samples is represented by the equation y = 131.3x-1.3172, with an R² value of 0.9945. Finally, the content of microplastics was determined by means of the calibration curve.

Serum hormone measurement

The concentrations of testosterone and follicle-stimulating hormone (FSH) in serum were measured using commercial ELISA kits (Ruixin Biotech, Quanzhou, China) according to the manufacturer’s instructions.

Cell culture

We purchased the mouse spermatocyte cell line, mycoplasma-free GC-2 spd (BALB/c, male), from the National Collection of Authenticated Cell Cultures (Shanghai, China) and successful authentication of the cell line was accomplished using STR Analysis. The cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco, Waltham, USA) supplemented with 10% (v/v) fetal bovine serum (FBS, Excell, Shanghai, China), penicillin (100 U/mL) and streptomycin (100 µg/mL) (Solarbio, Beijing, China) according to methods published previously [32]. The GC-2 cell line was cultured at 37 °C in a 5% CO₂ incubator. GC-2 cells were exposed by a variety of reagents, including 50, 100, 200 µg/mL PS-NPs, 100 µg/mL Cy5-PS, 3.5 µМ Deferiprone (DFP) (Glpbio, Montclair, USA), 10 mM 3-Methyladenine (3-MA) (Glpbio, Montclair, USA) and 5 µМ pioglitazone (Glpbio, Montclair, USA) according to specific experimental protocols.

Cell viability assay

Cell viability assay was performed according to our previous reported procedures with a few minor changes [32]. Cell Counting Kit-8 (CCK-8, Glpbio, Montclair, USA) was employed to ascertain the viability of the cells. After the cells were stimulated by different intervention factors, the culture medium was removed, and then the cells were incubated with the fresh medium supplemented 10 µL CCK-8 at 37 °C in a 5% CO₂ incubator for 3 h. Finally, the absorbance at 450 nm was quantified and employed to calculate the cell viability.

Furthermore, Calcein/PI Live/Dead Assay Kit (Beyotime, C2015L) were used to detect the cell survival rate. Briefly, after exposure, the cells were fluorescently labelled using a mixture of calcein-AM and propidium iodide (PI) solution for a duration of 30 min, and then the images were captured by the BX40 microscope. The cell survival rate was determined in 10 different fields.

Cellular uptake assay

To investigate the subcellular distribution of PS-NPs, the ability of cells to internalize PS-NPs was assessed by fluorescence microscope. Briefly, the cells were exposed with 100 µg/mL of Cy5-PS for an extra duration of 0–12 h. Following this treatment, the cells were imaged at specified time intervals (0, 4, 8, and 12 h) in vitro to monitor the fluorescence intensity of Cy5-PS in GC-2 cells.

Measurement of GSH and lipid peroxidation biomarker levels

The concentration of GSH in testes, serum, and cells was quantified using a GSH assay kit (Nanjing Jiancheng, A006-2, Nanjing, China). Similarly, the concentration of Malondialdehyde (MDA) in these samples was assessed using an MDA assay kit (Nanjing Jiancheng, A003-1). The protein concentration was gauged using the Enhanced BCA Kit (Beyotime, P0009). The concentrations of GSH and MDA were then calculated and adjusted according to the measured protein concentration, and subsequently normalized to the control level. Additionally, a fluorescence probe, C11-BODIPY 581/591 (Glpbio, GC40165, Montclair, USA) was employed to quantify the lipid peroxidation level. The images were captured by the microscope BX40, and the quantification of C11-BODIPY was performed on 10 different fields. For further quantification, Influx Cell Sorter (BD Biosciences, San Jose, CA, USA) was employed to conduct the Fluorescence-activated cell sorting (FACS) experiments, and then the data were analyzed with the FlowJo Software (San Jose, CA, USA).

Measurement of total Fe in testes and serum

The concentration of total Fe in testes and serum was quantified utilizing a total iron assay kit (Servicebio, Wuhan, China). Briefly, the procedure involved the homogenization of the isolated testes in PBS, followed by the collection of the supernatant post-centrifugation. Finally, the total Fe levels in the supernatant and serum were measured.

Cell and mitochondrial ROS assay

After GC-2 cells were treated as described above, the cells were stained with the fluorescence probe DCFH-DA (Beyotime, S0033) and MitoSOX (Glpbio, GC68230) and imaged with a microscope BX40. Additionally, the Influx Cell Sorter and FlowJo software were used to analyze the cellular and mitochondrial ROS.

Mitochondrial content assay

Mito-tracker Green (Beyotime, C1048) was utilized for incubation at 37 °C for 30 min in the dark, followed by the cell nuclei were fluorescently labelled with Hoechst 33342 (Beyotime, C1028). Images were captured using the fluorescence microscope BX40 for evaluation of the mitochondrial content.

Mitochondrial membrane potential (ΔΨm) assay

After the treatment as described above, the changes in ΔΨm were observed following labelling with a fluorescence probe TMRE (Beyotime, C2001). Furthermore, the cell nucleus was labelled with Hoechst 33342, and graphics were then captured using a fluorescence microscope.

Immunofluorescence assay

After treatment as described above, the cells were fixed with 4% paraformaldehyde at room temperature for a duration of 10 min. Subsequently, they were blocked with blocking buffer (Beyotime, P0102) at room temperature for a period of 1 h followed by incubation with FTH1 (Affinity, DF6278, 1:200), LC3α/β (Santa Cruz, sc-398822, 1:200) and NCOA4 (Santa Cruz, sc-373739, 1:100) at 4 °C overnight. Then, these cells were rinsed with PBST three times and then soaked with either fluorescent labeled Goat Anti-Mouse Secondary Antibody (Proteintech, RGAM002, 1:500) or Goat Anti-Rabbit Secondary Antibody (Proteintech, RGAR004, 1:500) at room temperature for a duration of 1 h.

The testicular sections embedded in paraffin were immersed in hematoxylin for 2-min, followed by a 1-min water rinse. The sections were blocked with commercialized blocking buffer at room temperature for 1 h, after which they were incubated with GPX4 (Affinity, DF6701, 1:250) and SLC7A11 (Affinity, DF12509, 1:250) in blocking buffer at 4 °C overnight. Subsequently, the cells were rinsed with PBST three times and incubated with fluorescent labeled Secondary Antibody (Proteintech, RGAR002, 1:500) at room temperature for a period of 1 h. The cell nuclei were labelled with DAPI (Beyotime, C1005).

Quantitative reverse-transcription PCR (qRT-PCR)

Total RNA was extracted from GC-2 cells or testes by utilizing the TRNzol reagent (Tiangen, China) and reverse transcribed into cDNA with SuperMix for qRT-PCR (+ gDNA wiper) (Vazyme, China) according to previous publications [32]. Nanodrop 2000 (Thermo, USA) was performed to measure RNA and cDNA concentrations, as well as their OD260/OD280 ratios. To optimize the use of RNA and cDNA, they were preserved at -80 °C. Subsequently, qRT-PCR was conducted on a PCR instrument LineGene 9600 Plus (Bioer, FQD-96 A) and ChamQ SYBR qPCR Master Mix (Vazyme, China) in a 20 µL reaction volume to determine the relative expression levels of the genes. The expression of β-actin was employed as an internal control to standardize the expression levels of target mRNA. The quantification of relative mRNA expression was performed utilizing the 2^−ΔΔCt method. Primers utilized for qRT-PCR are detailed in Table S1.

Western blotting

Western blotting was performed based on a previous publication with minor modifications [32]. Ice-cold RIPA lysis buffer (Beyotime, P0013B) supplemented 1 mM protease inhibitors (Solarbio, P6730, 1:100) and PMSF (Beyotime, ST506) was employed to extract the total protein from GC-2 cells and testes. The protein samples (20 µg) were resolved on 12% SDS-PAGE gels and subsequently electrophoretically transferred to polyvinylidene difluoride (PVDF, Millipore) membranes. The membranes were blocked with blocking buffer (Beyotime, P0252) at room temperature for 1 h and then soaked with primary antibodies, including those against LCN2 (Affinity, DF6816, 1:1000), NCOA4 (Affinity, DF4255, 1:1000), ACSL4 (Affinity, DF12141, 1:2000), GPX4 (Affinity, DF6701, 1:1000), SLC7A11 (Affinity, DF12509, 1:1000), FTH1 (Affinity, DF6278, 1:1000), STEAP3 (Affinity, AF0652, 1:1000), SQSTM1 (Affinity, AF5384, 1:1000), LC3 (Abcam, ab192890, 1:2000), CISD1 (Proteintech, 16006-1-AP, 1:5000), GAPDH (Proteintech, 60004-1-Ig, 1:50000), β-actin (Proteintech, 66009-1-Ig, 1:20000), followed by incubation at 4 °C overnight. After rinsing with TBST three times, the membranes were incubated at room temperature for 1 h either with HRP-conjugated anti-rabbit secondary antibody (Proteintech, SA00001-1, 1:5000) or anti-mouse secondary antibody (Proteintech, SA00001-2, 1:5000). The bands in membranes were detected using ChemiDoc XRS⁺ (Bio-Rad) after an instantaneous incubation was performed with enhanced ECL (Millipore, WBKLS0100). ImageJ software and Image Lab Software (Bio-Rad) were used to perform the semi-quantitative analysis. Relative protein expression levels were then standardized using either β-actin or GAPDH as appropriate for the different proteins.

Transmission electron microscopy (TEM) assay

TEM analysis was conducted to capture the subcellular structure and morphology of PS-NPs-treated GC-2 cells, as well as the subcellular localization of PS-NPs in GC-2 cells. In brief, GC-2 cells were exposed with PS-NPs for 12 h. Prior to cell collection, the cells were rinsed with PBS twice and then fixed in 4% glutaraldehyde at room temperature for a duration of 2 h. Subsequently, a graded series of alcohol and acetone was performed to dehydrate the cells, followed by embedding with HistoCore Arcadia (Leica, USA). The EMUC7 ultrathin microtome (Leica, USA) was used to prepare the ultrathin slides, and uranyl acetate and lead citrate were employed to stain these sections. Finally, the ultrathin sections were then imaged using a TEM.

RNAi and gene transfection

The mouse-specific NCOA4 siRNA (si-NCOA4), CISD1 overexpression (oe-CISD1), and NOCA4 overexpression (oe-NCOA4) were synthesized by and procured from GenePharma (Shanghai, China). si-NCOA4, oe-CISD1, and oe-NCOA4 were transfected into GC-2 cells using GP-Transfect-Mate (GenePharma, China), and these siRNA sequences are presented in Table S2. Briefly, siRNA (50 nM) and plasmid (50 nM) were transfected into cells for 24 h and 6 h, respectively, at 37 °C in a 5% CO₂ incubator for the next step of the experiment.

Detection of cellular and mitochondrial Fe²⁺ level

FerroOrange (Dojindo, F374, Japan), a fluorescent probe that enables the visualization of Fe²⁺, was employed to assess the intracellular Fe²⁺ and the mitochondrial Fe²⁺ was detected by a fluorescent probe Mito-FerroGreen (Dojindo, M489, Japan). After incubation for 30-min, cellular and mitochondrial Fe²⁺ in GC-2 cells were labelled with these probes and was observed using a fluorescence microscope BX40.

Proteome analysis

Proteome analysis was conducted on three biological replicates of GC-2 cells, yielding a total of six samples. These samples were rinsed with PBS prior to undergoing mass spectrometry, following the protocol detailed in previous methods [34]. The DEqMS method was employed for the proteome analysis [35]. Only proteins exhibiting P < 0.05 and|log₂ fold change| >1.2 were deemed significantly altered.

Statistical analysis

All experiments in this study were conducted at least three times. Paired Student’s t-tests were employed to ascertain whether there is a significant difference between the means of the two groups. One-way ANOVA followed by LSD test was performed to ascertain significant main effects using GraphPad Prism 8 software. The F-test included culture replicates as a random variable. In the event that the data were not normally distributed, as assessed by the Shapiro–Wilk test, they were converted into logarithms. The data are presented as means ± S.E.M. P < 0.05 or P < 0.01 were considered statistically significant. Additionally, the different letters indicate the existence of statistically represented significant differences (P < 0.05).

Results

Characterization of PS-NPs

TEM was employed to determine the size and morphology of PS-NPs, and the results demonstrated that the size of PS-NPs is 46.38 ± 3.94 nm, and the morphology of PS-NPs is spherical (Figure S1A). In addition, ZEN3600 was utilized to evaluate the stability of PS-NPs, and the results showed that the hydrodynamic diameter in the culture medium (DMEM, high glucose, Gibco, USA) was 48.02 ± 3.69 nm (Figure S1B), and that in the ddH₂O was 50.07 ± 1.10 nm (Figure S1D). The zeta potential value was − 52.53 ± 3.07 mV and − 42.61 ± 3.07 mV in culture medium and ddH₂O, respectively (Figure S1C and E). Together, these PS-NPs are applicable for subsequent experiments.

PS-NPs induced testicular damage and spermatogenesis disorder in mice

To explore the impact of PS-NPs on male reproductive system, we performed a PS-NPs-exposed mouse model by administering a PS-NPs suspension (10 mg/mL) via oral gavage, with a dose of PS-NPs was 50 mg/kg per day. As shown in Fig. 1A, exposure to PS-NPs resulted in non-significant change in the body weight of mice. However, it was noted that the cumulative content of PS-NPs in the mouse testes reached 12.08 µg/g (Fig. 1B). Although the exposure of PS-NPs had no significant effect on testicular size, weight and testes relative weight (the testes weight to body weight ratio) (Fig. 1C-E), histological structure of the testes of PS-NPs-exposed mice revealed that enlargement of seminiferous tubule lumen area, disruption of spermatocyte and conspicuous loss of spermatozoa were spotted in the PS-NPs group (Fig. 1F). Meanwhile, Johnsen’s score of testes in the exposure of PS-NPs was also significantly decreased (Fig. 1G). Consistently, the results of CASA indicated that there was an significant reduction of sperm quality, which was evident in terms of the viability (Fig. 1I), motility (Fig. 1J), and abnormality (Fig. 1K-L) of sperm. Although not statistically significant, a reduction in sperm concentration was observed (Fig. 1H). Given that spermatogenesis is a hormone-dependent process, we detected serum testosterone and follicle-stimulating hormone (FSH) concentrations, and these results demonstrated that PS-NPs markedly reduced the concentrations of both hormones (Fig. 1M-N). Taken together, these findings demonstrated that PS-NPs caused damage on the mouse testes, particularly to the spermatocyte.

Fig. 1.

PS-NPs cause severe dysfunction in mice testes. (A) Effect of PS-NPs on body weight. Mice were administered 50 mg/kg of PS-NPs via oral gavage for 35 days once every day. (B) PS-NPs contents in mice testes after 35 consecutive days of exposure to PS-NPs. (C) Representative photographs of isolated testes in mice after treatment. (D and E) Effect of PS-NPs on the testes weight and testes relative weight of mice. (F) Morphological observation of testes. The H&E-stained testes were visualized and the damaged spermatocyte in the seminiferous tubule are pointed by stars. (G) Johnsen Score of mice testes. (H-K) Effect of PS-NPs on sperm concentration, sperm viability, sperm motility and sperm abnormality. (L) Representative images of sperm morphology. The Abnormal sperm is indicated by arrows. (M and N) Testosterone and FSH concentration in serum at day 35 after treatment were assayed

PS-NPs caused ferroptosis of testes in mouse

Iron plays a crucial role in maintaining testosterone synthesis and spermatogenesis. To investigate the effects of PS-NPs on the testes, it is not adequate to rely solely on alterations in serum-related indicators. This is because both MDA and GSH content in the testes can be transported into the serum via blood circulation. Consequently, this study concurrently measured the levels of MDA, GSH, and total iron content in both serum and testes. The results indicated that the treatment of PS-NPs did not change the serum total Fe (Fig. 2A) and GSH contents (Fig. 2C), but increased the content of serum MDA (Fig. 2B). Meanwhile, PS-NPs induced a significantly higher in testicular total Fe (Fig. 2D) and MDA contents (Fig. 2E), and a significantly lower in testicular GSH level (Fig. 2F). Moreover, PS-NPs upregulated the mRNA expression of Ncoa4 but downregulated the mRNA expression of Gpx4 (Fig. 2G). Additionally, western blotting revealed that PS-NPs reduced ferroptosis-related proteins GPX4 and SLC7A11 (Fig. 2H-I) expression, which was also confirmed by the results of immunofluorescence (Fig. 2J-K). Taken together, our results demonstrated that PS-NPs triggered testicular ferroptosis in mice.

Fig. 2.

PS-NPs induce testicular ferroptosis in mice. (A-C) The concentration of total Fe, MDA and GSH in mice serum after PS-NPs exposure. (D-F) Effect of PS-NPs on total Fe, MDA and GSH in mice testes. (G) Gene expression of ferroptosis-associated genes in isolated testes using qRT-PCR after treatment. (H and I) Western blotting and quantitative analysis of ferroptosis-related protein expression. (J and K) Immunofluorescence images of GPX4 and SLC7A11 in testes following PS-NPs treatment for 35 days

PS-NPs-induced iron-overload led to ferroptosis in GC-2 cells

In order to further determine the impact of PS-NPs exposure on mouse testes, GC-2 cell line was selected in this study according to the results above and was exposed to PS-NPs to explore the cytotoxicity on spermatocyte. It was discerned that there was a marked dose-dependent diminishment for the viability of GC-2 cells after PS-NPs exposure for 12 h (Fig. 3A). In addition, the co-staining of calcein AM and PI underscored the toxicity of 100 µg/mL PS-NPs exposure on the cells for 12 h (Figure S2A), therefore, this treatment condition was selected for subsequent experiments. The results of in vivo experiments substantiated that PS-NPs caused damage to spermatocyte and induced ferroptosis in testes, and we speculated that ferroptosis occurred in spermatocyte. To test this hypothesis, the level of lipid peroxidation and the content of iron-related biochemical indicators were measured. Exposure of PS-NPs to GC-2 cells resulted in an increase in lipid ROS, as assayed using the fluorescence probe C11 BODIPY 581/591 (Fig. 3B-C). Similar to the results observed in testes, the MDA level was dramatically increased (Fig. 3D) and GSH level was significantly decreased (Fig. 3E) in GC-2 cells. Another feature of ferroptosis is iron overload. We also determined whether iron was required for the action of PS-NPs in GC-2 cells. It was observed that an elevating content of total Fe (Fig. 3F) and the increase in FerroOrange-sensitive Fe²⁺ (Fig. 3G-H) in PS-NPs-exposed GC-2 cells. In addition, we determined that GC-2 cells exposed to PS-NPs exhibited increasing ROS production, as the results detected by fluorescence microscope and flow cytometry using the fluorescence probe DCFH-DA (Fig. 3I-K). Subsequently, we quantified the levels of ferroptosis-related genes and proteins. The mRNA expression level of Ptgs2 (Fig. 3L) were significantly increased and the protein level of GPX4 were decreased (Fig. 3M-N). Mitochondria play a key role in ferroptosis induced by environmental pollutants. The TEM observation demonstrated that PS-NPs caused mitochondrial membrane rupture and mitochondrial cristae loss in GC-2 cells (Fig. 3O), accompanied with reduced mitochondrial abundance (Fig. 3P-Q) and a markedly downregulated mitochondrial membrane potential (Fig. 3R-S). Thus, these results indicated that PS-NPs could induce lipid peroxidation, iron-overload, and mitochondrial dysfunction, ultimately triggering ferroptosis of GC-2 cells.

Fig. 3.

PS-NPs initiate ferroptosis by triggering iron overload and ROS overproduction in GC-2 cells. (A) Various concentrations of PS-NPs (50 µg/mL, 100 µg/mL, 200 µg/mL) in different treatment times (6–24 h) reduce cell viability in GC-2 cells. (B and C) Representative images and quantification of fluorescence intensity for lipid peroxidation levels labelled with C11 BODIPY in GC-2 cells post-exposed with PS-NPs. Statistical analysis presented the ratio of MFI of green to red. (D-F) The levels of MDA, GSH, and total Fe were detected in GC-2 cells after PS-NPs treatment for 12 h. (G and H) Representative images and quantification of fluorescence intensity showing intracellular chelatable iron in GC-2 cells post-exposure to PS-NPs labelled with FerroOrange (red). (I and J) DCFH-DA staining was performed to detect ROS levels of GC-2 cells after PS-NPs exposure for 12 h using fluorescent microscopy. (K) Flow cytometry analysis of and ROS levels of GC-2 cells after PS-NPs exposure for 12 h using Fluorescence-activated cell sorting (FACS). (L) qRT-PCR analysis of ferroptosis-related gene expression in GC-2 cells treated with or without PS-NPs. (M and N) Ferroptosis-related proteins in GC-2 cells stimulated by PS-NPs were determined by western blotting. And semi-quantification of protein expression levels normalized to β-actin. (O) Representative TEM images depicting mitochondria in GC-2 cells in response to exposure to PS-NPs. The reduction of mitochondrial cristae is denoted by yellow arrows, while red arrows signify mitochondrial membrane rupture. (P and Q) Representative illustration and quantification of fluorescence intensity depicting mitochondria in GC-2 cells after treatment with PS-NPs using Mito-Tracker Green. (R and S) Illustration and quantification data of mitochondrial membrane potential in GC-2 cells in response to treatment of PS-NPs stained with TMRE

DFP ameliorated PS-NPs-triggered ferroptosis of GC-2 cells through iron chelation

In order to verify that PS-NPs caused iron overload and led to ferroptosis in GC-2 cells, we next investigated whether DFP, an iron chelator, could ameliorate PS-NPs-triggered ferroptosis of GC-2 cells. In agreement with our hypothesis, the cell viability had a significant rise after the co-treatment of both DFP and PS-NPs, compared to those of PS-NPs treated (Fig. 4A). As shown in Fig. 4B-C, DFP reduced the increase in lipid peroxidation in the co-treatment with DFP and PS-NPs. To further confirm the relationship between lipid peroxidation and DFP-chelated iron, MDA and GSH assays were performed, and the results showed that these biochemical markers were reversed by co-treatment of DFP and PS-NPs (Fig. 4D-E). Next, we detected the contents of total Fe and Fe²⁺ following exposure of PS-NPs with DFP treatment, which showed that the total Fe content decreased in the co-treatment with DFP and PS-NPs (Fig. 4F). Correspondingly, DFP reversed the rise in the concentration of Fe²⁺ caused by PS-NPs in GC-2 cells (Fig. 4G-H). Given that the increase of iron can trigger oxidative stress, which drives the cellular ferroptosis, therefore, we explored the relationship between DFP treatment and ROS production. The results indicated that, to some extent, DFP alleviated excessive production of ROS (Fig. 4I-K), but did not restore it to the normal level. Interestingly, the findings indicated that DFP mitigated the mitochondrial damage induced by PS-NPs. TEM observation showed that DFP alleviated the rupture of mitochondrial membrane and reduction of mitochondrial cristae in PS-NPs-treated GC-2 cells (Fig. 4L). In addition, the results showed that DFP relieved reduction of mitochondria caused by PS-NPs (Fig. 4M-N), and we also found that the level of mitochondrial membrane potential was improved with DFP and PS-NPs treatment (Fig. 4O-P), suggesting DFP limited mitochondrial dysfunction effectively. In aggregate, we demonstrated that DFP decelerated PS-NPs-induced ferroptosis in GC-2 cells through iron chelation.

Fig. 4.

DFP mitigates PS-NPs-induced ferroptosis in GC-2 cells. (A) The cell viability was evaluated in GC-2 cells in response to PS-NPs and DFP treatment. (B and C) Representative fluorescent images and quantification of C11 BODIPY in GC-2 cells after PS-NPs treatment with DFP. (D-F) The levels of MDA, GSH and total Fe were detected in GC-2 cells after PS-NPs treatment with DFP. (G and H) Intracellular chelatable iron in GC-2 cells following PS-NPs treatment with or without DFP stained with FerroOrange. And the fluorescence intensity was quantitatively analyzed. (I) The levels of ROS were detected using DCFH-DA (green) by FACS. (J and K) The ROS level were detected using the fluorescent indicator DCFH-DA in GC-2 cells following PS-NPs and DFP co-treatment using fluorescent microscopy. (L) Representative TEM images depicting mitochondria in GC-2 cells post-treatment with PS-NPs and DFP are presented. The reduction of mitochondrial cristae is denoted by yellow arrows, while red arrows signify mitochondrial membrane rupture. (M and N) Representative images and quantification of of Mito-Tracker-labelled mitochondria in GC-2 cells in response to PS-NPs and DFP co-treatment. (O and P) Images and quantification of fluorescence intensity of mitochondrial membrane potential in GC-2 cells stimulated by PS-NPs and DFP using the fluorescent probe TMRE

PS-NPs activated ferritinophagy by promoting the combination of NCOA4 and FTH1 in GC-2 cells

To determine the mechanisms underlying PS-NPs-induced ferroptosis in GC-2 cells, proteomic analysis was performed in this study. GO enrichment analysis revealed that autophagy, cellular iron ion homeostasis, iron-sulfur cluster binding, and other processes were regulated by PS-NPs (Fig. 5A). Heatmaps indicated that PS-NPs altered the expression of ferroptosis- (Fig. 5B) and autophagy-related proteins (Fig. 5C), such as FTH1, Ferritin Light Chain (FTL1), Autophagy Related 3 (ATG3), and Sequestosome 1 (SQSTM1). Next, the results of immunofluorescence analysis confirmed PS-NPs triggered autophagy in GC-2 cells, which was reflected in the increase of LC3 (Fig. 5D). Interestingly, the level of FTH1 protein was decreased after treatment with PS-NPs in GC-2 cells (Fig. 5F-G and S3A-B), yet its mRNA expression exhibited the opposite trend (Fig. 5E), suggesting that the reason for the reduction of FTH1 protein in PS-NPs-exposed GC-2 cells was not alteration of the gene transcription. Interestingly, exposure to PS-NPs increased the co-localization of NCOA4 and FTH1 for ferritinophagy (Fig. 5H). In addition, 3-MA, an autophagy inhibitor, effectively alleviated the reduced cell viability (Fig. 5I) and reversed the expression of autophagy- and ferroptosis-related proteins (Fig. 5J-L and S3C-D) changed by PS-NPs. Meanwhile, the elevated combination of NCOA4 and FTH1 caused by PS-NPs in GC-2 cells was inhibited by the co-treatment of 3-MA and PS-NPs (Fig. 5M). In order to elucidate the function of NCOA4 in PS-NPs-triggered ferritinophagy, NCOA4 in GC-2 cells was knocked down using NCOA4 siRNA (si-NCOA4), and we confirmed its silence by western blotting (Figure S4A-B). Notably, si-NCOA4 alleviated PS-NPs-reduced GC-2 cell viability (Fig. 5N), and mitigated PS-NPs-induced the upregulation of lipid peroxidation levels (Fig. 5O-P), MDA accumulation (Figure S5A), and GSH depletion (Figure S5B). Similarly, we found that NCOA4 knockdown effectively decreased the levels of total Fe, Fe²⁺ and ROS in PS-NPs-treated cells (Fig. 5Q-T and S5C). Together, these findings indicated that PS-NPs elevated free iron and lipid peroxidation levels by driving FTH1 degradation via NCOA4-mediated autophagy, ultimately leading to ferroptosis of GC-2 cells.

Fig. 5.

3-MA and NCOA4 knockdown suppress PS-NPs-induced ferritinophagy. (A) GO analysis based on proteomic analysis results of PS-NPs-treated group and control group. (B and C) Heat maps of ferroptosis-related and autophagy-related proteins in vehicle and PS-NPs-treated GC-2 cells. (D) Immunofluorescence of LC3 in GC-2 cells treated with PS-NPs. (E) Relative expression of ferritinophagy-related gene in GC-2 cells post-treatment PS-NPs. (F and G) Ferritinophagy-related protein expression in PS-NPs-stimulated GC-2 cells. (H) Immunofluorescence of NCOA4 and FTH1 in GC-2 cells treated with PS-NPs. Line intensity plots show colocalization between FTH1 (red) and NCOA4 (green). (I) The cell viability was assayed in GC-2 cells following PS-NPs treatment with or without 3-MA. (J) Immunofluorescence staining of LC3 in GC-2 cells post-treatment with PS-NPs and 3-MA. (K and L) Western blotting and quantitative analysis of NCOA4 and FTH1. (M) Immunofluorescence of NCOA4 and FTH1 in GC-2 cells following PS-NPs and 3-MA treatment. Line intensity plots show colocalization between FTH1 (red) and NCOA4 (green). (N) The viability of PS-NPs-stimulated GC-2 cells while simultaneously inhibiting NCOA4 expression. (O and P) Representative fluorescent images and quantitative analysis of lipid peroxidation levels in PS-NPs-stimulated GC-2 cells while simultaneously inhibiting NCOA4 expression. (Q and R) Intracellular chelatable iron in GC-2 cells after 12 h exposure to PS-NPs while inhibiting NCOA4 expression stained with FerroOrange. (S and T) Representative fluorescent images and quantification of ROS levels using the fluorescent indicator DCFH-DA in GC-2 cells after 12 h exposure to PS-NPs while inhibiting NCOA4 expression

Internalized PS-NPs accumulated in mitochondria and induced dysfunction in GC-2 cells

To investigate the intrinsic reasons for cytotoxicity caused by PS-NPs, we next investigated the subcellular distribution of PS-NPs in GC-2 cells. The cells were exposed with Cy5-PS for 12 h, and the results showed that Cy5-PS could be internalized into GC-2 cells and clustered around the nucleus (Fig. 6A). Similarly, the results of TEM indicated that PS-NPs were internalized into GC-2 cells and accumulated successively in lysosomes and mitochondria (Fig. 6B and Figure S6). To investigate whether the accumulation of PS-NPs is time-dependent, the co-localization of Cy5-PS with lysosome and mitochondria was detected and the results showed that a significantly weaker overlap of Cy5-PS and Lyso-tracker (lysosome) (Fig. 6C), while a significantly stronger overlap of Cy5-PS and Mito-tracker (mitochondria) ranged from 4 to 12 h (Fig. 6D) were observed. To further investigate the degree of mitochondrial damage caused by PS-NPs internalized into GC-2 cells, changes in mitochondrial Fe²⁺ and mitochondrial ROS (mtROS) levels was detected. We found that the fluorescence intensity of Mito-FerroGreen in PS-NPs-treated cells was increased, indicating that PS-NPs exposure resulted in an increased mitochondrial Fe²⁺ level in GC-2 cells (Fig. 6E-F). In addition, mtROS levels were measured using flow cytometry, and the results exhibited that PS-NPs induced excessive production of mtROS (Fig. 6G). These findings showed that PS-NPs preferentially aggregated in the lysosome and subsequently accumulated in mitochondria, resulting in dysfunction by inducing an increased mitochondrial Fe²⁺ level and an elevation of mtROS level in GC-2 cells.

Fig. 6.

PS-NPs enter cells via endocytosis, subsequently reaching the mitochondria and causing mitochondrial damage. (A) Distribution of Cy5-NPs (red) in GC-2 cells detected by fluorescent microscopy, which overlapped with nucleus (blue) and cytoplasm (green). (B) Representative TEM images of PS-NPs in cells. The red arrow in a’ represents PS-NPs that are entering the cell through endocytosis. b’ refers to abundant accumulation of PS-NPs in lysosomes. The blue arrows in c’ refer to that PS-NPs swell the lysosome and are released from it. The red circles indicate PS-NPs localized within mitochondria, which were initially released from lysosomes into the cytoplasm before ultimately accumulating in mitochondria. (C) The images demonstrate the time-dependent co-localization of Cy5-NPs (red) and lysosomes (green) in GC-2 cells. Line intensity plots show colocalization between Cy5-NPs (red) and lysosomes (green). (D) The images of the time-dependent co-localization of Cy5-NPs (red) and mitochondria (green) in GC-2 cells. Line intensity plots show colocalization between Cy5-NPs (red) and lysosomes (green). (E and F) Representative images and quantification of Fe²⁺ of mitochondria in GC-2 cells after 12 h exposure to PS-NPs. (G) Detection of mitochondrial ROS in GC-2 cells treated with PS-NPs for 12 h

CISD1 downregulated by PS-NPs aggravated the import of Fe²⁺ released by ferritinophagy into the mitochondria of GC-2 cells

To clarify the relationship between ferroptosis and mitochondrial dysfunction in GC-2 cells induced by PS-NPs, the expression and the function of CISD1 were investigated according to the results of proteomic analysis (Fig. 5B). The western blotting and immunofluorescence for CISD1 results confirmed that PS-NPs induced a decline of CISD1 (Fig. 7A-B and S7A-B). The CISD1 overexpression vector (oe-CISD1) was synthesized and transfected into the GC-2 cells for rescue experiments. The western blotting results confirmed that transfection of oe-CISD1 upregulated CISD1 expression (Fig. 7C-D). As expected, CISD1 overexpression significantly ameliorated the decrease of GC-2 cell viability triggered by PS-NPs (Fig. 7E), lipid peroxidation acceleration (Fig. 7F-G), MDA content (Figure S8A), and ROS excess (Figure S8D-E). Of note, we found that oe-CISD1 did not significantly affect the depletion of GSH (Figure S8B), total Fe (Figure S8C), and Fe²⁺ (Fig. 7H-I) accumulation in GC-2 cells following treatment with PS-NPs, while it downregulated the mitochondrial Fe²⁺ caused by PS-NPs (Fig. 7J-K). Collectively, these findings suggested that the downregulation of CISD1 was an important contributor to ferroptosis triggered by PS-NPs.

Fig. 7.

PS-NPs regulate ferroptosis in GC-2 cells through NCOA4-mediated ferritinophagy-CISD1 axis. (A and B) CISD1 protein level in GC-2 cells stimulated by PS-NPs for 12 h. (C and D) CISD1 protein level in GC-2 cells transfected with oe-CISD1. (E) The viability of PS-NPs-treated GC-2 cells with CISD1 overexpression. (F and G) The images demonstrate the effect of PS-NPs treatment on lipid peroxidation levels in GC-2 cells with CISD1 overexpression, using a fluorescent probe C11 BODIPY to stain the cells. (H and I) Intracellular chelatable iron in PS-NPs-treated GC-2 cells with CISD1 overexpression using a fluorescent probe FerroOrange. (J and K) Representative images and quantification of mitochondrial Fe²⁺ in GC-2 cells exposure to PS-NPs with CISD1 overexpression. (L and M) Western blotting of NCOA4 protein levels in GC-2 cells transfected with oe-NCOA4. (N) Cell viability was assayed in pioglitazone-treated GC-2 cells with NCOA4 overexpression. (O and P) The images demonstrate the effect of pioglitazone treatment on lipid peroxidation levels in GC-2 cells with NCOA4 overexpression. (Q and R) Intracellular chelatable iron in pioglitazone-treated GC-2 cells with NCOA4 overexpression. (S and T) Representative images and quantification of mitochondrial Fe²⁺ in GC-2 cells exposure to pioglitazone with NCOA4 overexpression

To explore the relationship between ferritinophagy and CISD1 in this study, according to the above results, we supposed that the downregulation of CISD1 enhanced mitochondrial uptake of Fe²⁺ released by NCOA4-mediated ferritinophagy. To test this hypothesis, NCOA4 overexpression vector (oe-NCOA4) was synthesized and employed to transfect into the GC-2 cells. The results of western blotting showed successful overexpression of NCOA4 in GC-2 cells (Fig. 7L-M). It has been confirmed that pioglitazone increases the stability of CISD1 and suppresses the transfer of iron from CISD1 to mitochondria [36]. In this study, our results indicated that pioglitazone relieved cell death (Fig. 7N), mitigated lipid peroxidation (Fig. 7O-P), reduced MDA (Figure S9A), Fe²⁺ (Fig. 7Q-R), total ROS (Figure S9D-E), and mitochondrial Fe²⁺ (Fig. 7S-T) accumulation in NCOA4-overexpressing GC-2 cells. Nevertheless, pioglitazone did not rescue the depletion of GSH (Figure S9B) and increase in total iron (Figure S9C) caused by NCOA4 overexpression. Together, our results demonstrated that CISD1 downregulated by PS-NPs aggravated the ferroptosis by promoting the absorption of Fe²⁺ released by ferritinophagy in the mitochondria of GC-2 cells.

Discussion

That MPs/NPs were detected in human blood, the placenta, liver and other tissues suggests they could accumulate in the body, raising widespread concerns regarding the associated risks of MPs/NPs [3, 4, 29]. Although their potential to cause the male reproductive toxicity has been extensively studied [30–32], the underlying mechanism of testicular injury caused by PS-NPs still needs to be clarified. In the present study, we found ferroptosis in PS-NPs-treated spermatocytes and in mice exposed with PS-NPs. We show that PS-NPs exposure induces testicular damage and disrupts spermatogenesis. Mechanistically, PS-NPs enter GC-2 cells via internalization and accumulate successively in lysosomes and mitochondria, regulated by NCOA4-mediated ferritinophagy and CISD1 downregulation, which aggravated the flows of ferrous iron from cytoplasm to mitochondria. These results corroborate that CISD1 is an underlying genetic predisposition to PS-NPs-induced ferroptosis.

The size of NPs determines their intrinsic toxicity, which in turn affects their uptake and distribution [41]. MPs were detected in testicular and seminal samples, with polystyrene MPs (PS-MPs) being the predominant type found in human testes [10]. Existing research showed that 120 nm nanoparticles accumulated and retained in the intestine [42]. Studies have demonstrated that PS-NPs could induce the injury of liver, lung, and brain, resulting in metabolic disorders and reduced reproductive capabilities [1, 35, 36]. In this study, we investigated the content of 50 nm PS-NPs and the effect of PS-NPs on testes in mice. In addition, our results showed a significant increase in the concentration of PS-NPs in testes following continuous oral gavage for 35 days. Although not statistically significant, a reduction in sperm concentration and other damage were observed [45] This suggests that PS-NPs accumulated in testes and caused damage to spermatocytes and sperm, indicating that PS-NPs accumulated on reproductive organs through blood circulation and led to damage to spermatocytes. Apart from these cells, Leydig cells and Sertoli cells represent two major targets of PS-NPs toxicity, leading to rupture of the blood-testis barrier and steroid hormone synthesis disorders [13, 32]. The progression of spermatogenesis is contingent upon the integrity of testicular tissue structure [46]. Although the testicular environment is complex, given the findings from HE staining and in vivo experiments, it is evident that damage to spermatocytes is substantial, which suggests that PS-NPs primarily affect spermatocytes, leading to testicular damage. To further elucidate the impact of PS-NPs on the testes, this study employed in vitro experiments with GC-2 cells, a mouse spermatocyte cell line, as the model. It has been reported that exposure to polystyrene MPs can disrupt spermatogenesis in rats and interfere with the hypothalamic-pituitary-gonadal axis [47]. A similar study found that PS-MPs exposure triggered the reduction of sperm quality in mice via the NRF2/HO-1/NF-κB pathway [48]. Furthermore, a previous study demonstrated that exposure to PS-MPs caused severe damage to ovaries and testes, suggesting that female mice may be more susceptible to the reproductive and fertility effects of MPs than their male counterparts. In this study, we found that the accumulation of PS-NPs on testes and PS-NPs treatment intensified decline in sperm quality. In mouse testes, exposure to MPs has been shown to have detrimental effects on male reproduction and sperm quality, suggesting a potential risk to reproductive success. Together, these results corroborate the previous studies that PS-NPs can cause testicular damage [25, 31, 32].

Although existing research has found damage to the reproductive system caused by NPs, their effects and mechanisms of damage are still unclear. In male reproduction, iron plays an important role in maintaining testosterone synthesis and spermatogenesis [49]. Ferroptosis is an iron-dependent cell death form driven by lipid peroxidation [15]. Recent studies have shown that ferroptosis acts as an emerging mechanism implicated in environmental contamination [41–43]. PS-NPs are known to induce male reproductive toxicity through Nrf2-dependent spermatogenic cell ferroptosis [53], which is similar to our results showing ferroptosis in the mouse testes and spermatocytes upon exposure to 50 nm PS-NPs. There are several alterations in PS-NPs-treated GC-2 cells, including accumulation of Fe²⁺ and lipid ROS, increased MDA contents and GSH depletion. It is worth noting that PS-NPs treatment decreased the expression of GPX4 in mice testes. Recent studies found that the regulation of SLC7A11/GPX4 pathway is one of core molecular mechanisms of ferroptosis [45–47]. Additionally, GPX4, as the most critical selenoenzyme in testes, protects sperm from DNA damage caused by oxidative stress during early developmental stages, highlighting its importance in PS-NPs-induced testicular injury [57]. However, our results showed that the expression of GPX4 was not reduced by PS-NPs in GC-2 cells, suggesting that GPX4 is not the key protein involved in PS-NPs-induced ferroptosis in GC-2 cells, thereby adding complexity to the interpretation. We speculate that the deletion of GPX4 occurs in other cells, such as Leydig cells or Sertoli cells even spermatogonia, rather than spermatocytes because of the fact that testes is composed of various cell types. Thus, it is necessary for further study on its mechanism to screen the key proteins in GC-2 cells targeted by PS-NPs.

The primary cause of iron accumulation following PS-NPs treatment is the degradation of ferritin. Cellular iron is predominantly stored in ferritin, which is composed of FTL and FTH1 subunits [15]. Notably, in the present study, the alteration of FTH1 protein level was opposite to its mRNA expression, indicating that reduction of FTH1 was not attributable to transcriptional changes in response to PS-NPs treatment. Previous studies have demonstrated that autophagy, particularly NCOA4-mediated ferritinophagy, is crucial for recognizing FTH1 and facilitating its lysosomal degradation via the utilization of NCOA4 as the primary cargo receptor [16, 49, 50]. Here, our results of proteomic analysis showed that differentially expressed proteins in GC-2 cells treated with PS-NPs were enriched in biological processes related to autophagy and ferroptosis, and we further demonstrate that DFP, an iron chelator, substantially alleviates PS-NPs-induced ferroptosis in GC-2 cells. The autophagic degradation of ferritin is largely regulated by the cellular levels of NCOA4 [60]. An unanticipated finding in the current study was that both the mRNA expression and protein level of NCOA4 remained unchanged, yet an increased colocalization of NCOA4 and FTH1 was observed, indicating that PS-NPs promoted the interaction between NCOA4 and FTH1. It is worth noting that PS-NPs-induced ferroptosis was suppressed by the autophagy inhibitor 3-MA and knockdown of NCOA4, providing the evidence that ferritinophagy is involved in PS-NPs-induced ferroptosis in GC-2 cells.

As we all know, endocytosis plays a fundamental role in the regulation of cellular functions [52, 53]. It is noteworthy that polystyrene MPs are the predominant type found in human testes, with a density of 1.05 g/cm³, which is similar to the densities of body fluids [10, 54]. This suggests that PS-NPs used in this study are able to effectively access and interact with spermatocytes rather than remaining on the cell surface, thereby facilitating their uptake by spermatocytes. The interaction between NPs and cells is an intricate process that determines the internalization fate of NPs via endocytosis pathways [64]. The internalization pattern of NPs is contingent upon the type of cell, components of the plasma membrane that regulate their dynamics, and the physicochemical attributes of NPs. A previous study demonstrated that RAW 264.7 cells employed an endosomal-lysosomal route for uptake, in contrast to the caveolar uptake mechanism utilized by BEAS-2B human normal lung epithelial cells [65]. Similarly, a recent study elegantly showed that 20 nm PS-NPs were time-dependently internalized by macrophages and accumulated in the mitochondria [44]. Consistent with these results, our results clearly showed that PS-NPs entered spermatocytes via pinocytosis and exhibited a time-dependent accumulation around the nucleus within 12 h. The subcellular localization of PS-NPs within key organelles plays a crucial role in their ultimate fate and cellular toxicity. Generally, internalized PS-NPs accumulate in lysosomes and cause lysosomal membrane rupture and dysfunction of other organelles [66]. Here, we present the dynamic pattern of PS-NPs distribution in GC-2 cells. After cell uptake, PS-NPs initially accumulated in lysosomes, leading to the rupture of the lysosome, and were then released to mitochondria, resulting in the excessive Fe²⁺ and ROS in mitochondria, which is consistent with the previous study [67]. The current discovery corroborates the findings of the previous work in human astrocytoma, where PS-NPs were distributed in different organelles over time [68]. Furthermore, similar effects have been reported with other NPs and lysosomotropic reagents, which could be explained with the so-called ‘proton sponge theory’. Nevertheless, additional pathways or mechanisms of accumulation of PS-NPs in mitochondria cannot be excluded.

Mitochondria, as organelles responsible for energy production, are instrumental in preserving cellular homeostasis and function [69]. However, mitochondria exhibit a high vulnerability to environmental toxins, rendering them susceptible to damage, such as NPs exposure. A strong association between NPs and mitochondrial toxicity, resulting in various forms of damage, has been well documented in previous studies [70, 71]. Moreover, CISD1, an outer membrane protein of mitochondria, inhibits uptake and transport of Fe²⁺ into the mitochondrial matrix [22]. Here, we elucidate the mechanism by which the PS-NPs activate NCOA4/CISD1-mediated ferroptosis in GC-2 cells. To our surprise, CISD1 expression was significantly downregulated after PS-NPs treatment in spermatocytes. We speculate that a portion of the PS-NPs releases due to lysosomal rupture causes ferritinophagy, while another portion reaches to mitochondria, resulting in a reduction of CISD1. This reduction enhances the transfer of Fe²⁺ in the mitochondria released by ferritinophagy, thereby impairing mitochondrial function. This hypothesis is corroborated by our subsequent studies. Moreover, pioglitazone, a drug that increases the stability of CISD1 [36], markedly rescued ferroptosis induced by overexpression of NCOA4. Collectively, the relationship between ferritinophagy and CISD1 emerges as a promising therapeutic target for diseases related to iron metabolism.

Although we have pinpointed CISD1 as a potential target for PS-NPs to induce spermatocyte ferroptosis, our current study is not without its limitations. First, considering the complex physicochemical characteristics of NPs, it is necessary to investigate whether PS-NPs with different properties exert different effects on spermatocytes. This includes exploring the role of protein corona, charge, and particle size in testicular injury and spermatogenesis disorder caused by PS-NPs-induced ferroptosis. Second, our study focused on the distribution of internalized PS-NPs in spermatocytes, rather than how to activate pinocytosis. This limitation highlights the necessity for further research to determine whether there is a specific recognition of protein corona and membrane proteins in spermatocytes. Third, the off-target effects might have existed given the sole use of CISD1 inhibitors, and further research should incorporate genetic approaches knockdown or knockout of CISD1 models and perform comparative analyses of their effects on male reproductive capacity in order to precisely delineate CISD1’s function in PS-NPs-triggered testicular ferroptosis. Lastly, it is important to note that other targets besides CISD1 may be implicated in PS-NPs-induced testicular injury, and research investigating various targets of PS-NPs is urgently required and eagerly awaited, which will contribute to a more comprehensive understanding of the molecular mechanisms underlying human reproductive diseases.

Conclusion

In conclusion, we demonstrated that PS-NPs triggered the damage to testes through the ferroptosis of spermatocytes regulated by NCOA4-mediated ferritinophagy and CISD1 downregulation, which aggravated the flow of ferrous iron from cytoplasm to mitochondria. These findings advance our comprehension of the male reproductive toxicity induced by NPs and bioaccumulation resulting from NPs exposure, and delineate potential avenues for subsequent research into the prevention and treatment of male reproductive disorders linked to PS-NPs.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

MPs

Microplastics

NPs

Nanoplastics

PS-NPs

Polystyrene nanoplastics

SLC7A11

Solute carrier family 7 member 11

GSH

Glutathione

ACSL4

Acyl-CoA synthetase long-chain family member 4

GPX4

Glutathione peroxidase 4

NCOA4

Nuclear receptor coactivator 4

FTH1

Ferritin heavy chain 1

CISD1

CDGSH iron sulfur domain 1

HIF-1

Hypoxia-inducible factor-1

SPF

Specific pathogen-free

DLS

Dynamic light scattering

CASA

Computer assisted sperm analysis

DFP

Deferiprone

3-MA

3-Methyladenine

MDA

Malondialdehyde

FACS

Fluorescence-activated cell sorting

FSH

Follicle-stimulating hormone

FTL1

Ferritin light chain

ATG3

Autophagy related 3

SQSTM1

Sequestosome 1

Author contributions

Zhongliang Jiang conceptualized and supervised the overall project. Jing Lv and Guangyu Liu performed the cell culture experiments, contributed to the initial manuscript draft and conducted statistical analysis. Ziqi Wang, Jueshun Zhang performed histological assessments and coordinated the animal treatments. Yuanyou Li handled data visualization. Jing Lv and Guangyu Liu also managed figure production. Yifan Wang and Ning Liu were responsible for in vivo data collection. Zhongliang Jiang and Shayakhmetova Altyn critically revised the manuscript. All authors reviewed and approved the final version.

Funding

This work was supported by grants from the National Key R&D Program of China (2024YFE0213400), the National Natural Science Foundation of China (32272880) and the Key Project of Research and Development Program in Shaanxi Province, China (2024NC-YBXM-088).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participle

All animal experiments were carried out in accordance with ethical approval from the Institutional Animal Care and Use Committee of Northwest A & F University (Approval No.: IACUC2024-0510).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.