Abstract Microplastics (MPs), as emerging contaminants, may adversely affect male reproductive health. This study investigated the potential association between MP accumulation in human semen and sperm quality. Furthermore, the molecular mechanisms underlying MPs-induced sperm quality impairment were characterized using representative polystyrene microplastics (PS-MPs), murine models, and spermatogonial cell cultures. Among 200 semen samples, the overall detection rate of MPs was 55.5% (111/200). A total of 128 MPs were identified in semen, with PS (32.03%) and PVC (36.72%) being the predominant microplastic polymers. Epidemiological analyses revealed a significant positive association between plastic tableware (PT) use frequency and MP accumulation in semen. Stratified analyses further revealed a strong association between total MPs exposure and reduced sperm concentration among individuals with BMI < 24 kg/m² and frequent PT use. In murine models, exposure to PS-MPs induced reduced sperm quality, elevated sperm abnormalities, and increased levels of autophagy and apoptosis. Mechanistically, PS-MPs activated the MAP3K1/p38/c-fos pathway via the transcription factor FOXA1, thereby inducing the autophagy and apoptosis of spermatogonia. Collectively, this study provides direct human evidence that MP accumulation in semen is associated with impaired sperm quality, particularly in individuals with certain lifestyle factors such as frequent PT use. Moreover, our findings further demonstrate the potential reproductive toxicity of MPs and, for the first time, elucidate the critical role of the FOXA1/MAP3K1/p38 cascade in PS-MPs-mediated decline in sperm quality. Graphical abstract [34]graphic file with name 12951_2025_3747_Figa_HTML.jpg Supplementary Information The online version contains supplementary material available at 10.1186/s12951-025-03747-7. Keywords: Microplastics, Polystyrene microplastics, Plastic tableware use, Sperm quality decline, FOXA1/MAP3K1/p38 cascade Introduction Globally, approximately 15% of couples experience infertility, with male factors contributing to 50% of cases. A key determinant of male infertility is the decline in semen quality. Historical data indicate that the average sperm count declined progressively by 1% annually during 1938–1990, with the rate of decrease accelerating to 2.64% per year from 2000 onward [[35]1]. Accumulating evidence implicates environmental chemical exposures as critical risk factors, alongside age, body mass index (BMI), and lifestyle [[36]2]. Among these environmental chemicals, microplastics (MPs)—small plastic fragments or particles less than 5 mm in diameter, which are mainly derived from the manufacture and use of plastics—are of particular concern. The United Nations Environment Programme (UNEP) has identified MPs as one of the top ten emerging global environmental issues. Preliminary studies, though limited, have shown that MPs could accumulate in mammalian testes and induce detrimental reproductive effects [[37]3]. Fang et al. reported that PS-MPs impaired sperm quality by inducing mitochondrial oxidative stress and apoptosis in spermatogonia [[38]4]. Our recent research also uncovered that exposure to 50 μm PS-MPs interfered with the hypothalamic-pituitary-testis axis, reducing testosterone levels in mice via GPX1-mediated endoplasmic reticulum stress [[39]5]. With the accelerating pace of modern life and the rapid development of the internet, takeout food (often packaged in plastic containers) has become a preferred option for many people, leading to a significant rise in the use of plastic tableware (PT). Therefore, investigating the potential health risks of PT has become increasingly imperative and meaningful. Plastic tableware is commonly made from materials such as polystyrene (PS), polyvinyl chloride (PVC), polyethylene (PE), polypropylene (PP), and polyethylene terephthalate (PET). An investigation showed that when pure water at 100 °C was poured into a takeaway plastic coffee cup and left for 20 min, trillions of microplastic particles could be detected per liter of water, with particle release being directly correlated with water temperature [[40]6]. Simulating typical takeaway scenarios and delivery conditions, researchers further tested four types of plastic tableware, namely PP, PS, PE, and PET. The results revealed that all plastic containers released MPs, with PS-MPs being the most abundant [[41]7]. Moreover, researchers observed that after volunteers were exposed to heat-treated PT for one month, they exhibited a significant increase in fecal MPs, accompanied by changes in gut microbiota diversity and alterations in metabolite levels [[42]8]. Although recent experimental studies have indicated that MPs may accumulate in human semen and testicular tissues, existing evidence remains limited and inconclusive [[43]2]. Additionally, the association between PT use frequency and sperm quality remains poorly characterized. To bridge epidemiological observations and biological causality, it is imperative to dissect the molecular pathways by which MPs impair spermatogenesis. Sperm quality regulation is an intricate and delicate biological process. Mounting evidence suggests that apoptosis and autophagy are critical regulators of sperm quantity and quality, but their dysregulation may contribute to male infertility [[44]9, [45]10]. The mitogen-activated protein kinase (MAPK) signaling pathway, comprising Ser/Thr protein kinases, is functionally active in germ cells. This conserved pathway encompasses four major subfamilies: extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK), p38, and ERK5. Among these, p38 is a well-characterized signaling molecule that plays an essential role in various aspects of spermatogenesis and sperm development, such as spermatogonial differentiation, apoptosis, capacitation, and autophagy [[46]11]. A previous study revealed that benzyl butyl phthalate (BBP) induced autophagy and apoptosis of mouse spermatogonia (GC-1 cells) by activating the p38 pathway [[47]12]. Given the established role of p38 in environmental toxicant-induced germ cell damage, we hypothesized that MPs might similarly activate this pathway to disrupt spermatogenesis. However, whether the p38 pathway plays a significant role in MPs-mediated male reproductive damage remains unclear and requires further validation. To determine the effect of frequent PT use and the role of the p38 pathway in sperm quality following MPs exposure, we first investigated the association between PT use frequency and MP levels in semen samples from reproductive-age men, and subsequently explored how exposure to MPs might affect semen quality. Based on epidemiological evidence and supporting literature, we selected the representative PS-MPs—one of the most prevalent MPs detected in human semen—for subsequent in vivo and in vitro studies. Our epidemiological analyses revealed significant correlations among PT use frequency, cumulative MP accumulation, and reduced sperm concentration. Mechanistic investigations demonstrated that exposure to 50 nm PS-MPs not only impaired semen quality, but also induced the FOXA1-mediated MAP3K1/p38 pathway in both mouse testes and spermatogonia, thereby leading to spermatogonial autophagy and apoptosis. Collectively, this research advances our understanding of PT-associated MP accumulation in semen and its detrimental effects on male reproductive health, while also providing mechanistic insights into the FOXA1/p38 pathway that could facilitate the development of targeted interventions. Materials and methods Study participants and semen analysis This study included 200 men of reproductive age who visited the Chongqing Human Sperm Bank between May 2020 and December 2021. Written informed consent was obtained from all participants. Each participant completed a questionnaire addressing demographic characteristics (age, household income, education level, reproductive history, etc.), lifestyle factors (smoking, drinking, PT use frequency, etc.), occupational exposures, and medical history. Semen samples were collected and analyzed for semen quality parameters in accordance with the 6th edition of the World Health Organization (WHO) standards [[48]13], with specific evaluations including sperm concentration, total sperm count, progressive sperm motility, total sperm motility, and sperm immobility. The study was approved by the Ethics Committee of the Chongqing Population and Family Planning Science and Technology Research Institute. Analysis of MPs in semen Semen samples were first lyophilized and then treated with KOH solution, followed by stirring at 45 °C for 48 h to facilitate lysis. Undigested particles were filtered out using a 10 μm pore-size filter membrane. The remaining particles were sequentially treated with SDS solution and cellulase for enzymatic hydrolysis, then digested with Fenton’s reagent, and finally filtered again. The type and quantity of microplastics (MPs) in the filtrate were determined using microscopic infrared (micro-IR) spectroscopy, and their average concentrations were calculated. The microstructure of various MPs was observed using Scanning Electron Microscopy (SEM). A multivariable linear regression model was employed to assess the association between MP concentrations and semen quality parameters. Stratified analyses were used to evaluate potential effect modification by age, BMI, and PT use frequency. Preparation of PS-MPs Monodisperse polystyrene microspheres (2.5% w/v PS-MPs) and fluorescent polystyrene microspheres (1.0% w/v PS-MPs, red) were purchased from Baseline ChromTech Research Center (Tianjin, China). Three sizes of PS-MPs (50 nm, 5 μm, and 50 μm) were mounted on silicon wafers for morphological characterization using SEM. For in vivo and in vitro experiments, PS-MPs were dispersed in sterile ddH[2]O and PBS, respectively, and then sonicated for 30 min to ensure uniform dispersion before use. Animals and treatments Forty-eight male BALB/c mice (21.40 ± 0.93 g) were purchased from HFK Bioscience (Beijing, China). All mice were housed under a 12 h light/dark cycle with food and water ad libitum. After one week of acclimatization, animals were randomly divided into four groups (n = 12/group), stratified by initial body weight to ensure group balance. Animals were gavaged with ddH[2]O (control group) or 50 nm, 5 μm, or 50 μm PS-MPs (1 mg/day) for 8 consecutive weeks. The gavage dose was determined based on human daily microplastic intake and calculated using body surface area normalization [[49]14, [50]15]. Animals were sacrificed under diethyl ether anesthesia within 24 h after the last dose. All experimental procedures were approved by the Institutional Animal Care and Use Committee of Chongqing Population and Family Planning Science and Technology Research Institute. Cell culture and treatments The mouse testicular spermatogonial cell line GC-1 was purchased from ATCC (Rockville, CT, USA) and cultured in DMEM supplemented with 10% fetal bovine serum (HAKATA, Huiying biotechnology Co. Ltd., China), 100 U/mL penicillin, and 100 µg/mL streptomycin. After 24 h of growth, GC-1 cells were pretreated with either 1 µM p38 inhibitor Adezmapimod (SB203580) for 1 h or FOXA1 siRNA (Sangon Biotech., Shanghai, China) for 4 h, followed by incubation with 50 nm PS-MPs (25 µg/mL) in serum-free DMEM for 24 h. Assessment of sperm quality After euthanasia, the harvested mouse caudal epididymides were placed in 0.5 mL PBS, minced with fine forceps, and incubated at 37℃ for 30 min. A 10 µL aliquot of suspension was transferred to a Makler counting chamber, and the sperm motility and concentration were analyzed under a light microscope. Each test was performed in duplicate, with 200 spermatozoa counted per animal. Additionally, 10 µL samples were placed on a glass slide, smeared, and stained using the modified Papanicolaou staining method. The slides were then mounted with a coverslip, and sperm morphological abnormalities were assessed under a light microscope. For quantitative analysis, five random fields of view were imaged per smear, and the sperm malformation rate was calculated for each experimental group. Accumulation of PS-MPs in mouse testes and spermatogonia Five-week-old BALB/c mice were gavaged daily with either ddH[2]O (control) or fluorescent PS-MPs (50 nm, 5 μm, and 50 μm) for one week. The biofluorescence imaging (BFI) system LB 983 NightOWL II (Berthold, Germany) was used to visualize the accumulation and distribution of PS-MPs in the stomach (positive control, as direct oral exposure ensures PS-MP presence in the stomach) and testicular tissues. The fluorescent PS-MPs were observed at Ex: 543 nm and Em: 605 nm. To investigate cellular uptake, GC-1 cells (1 × 10^6 cells per well) were seeded in a 6-well plate and incubated overnight. Cells were subsequently exposed to 25 µg/mL fluorescent PS-MPs (50 nm, 5 μm, and 50 μm) at 37 °C for 10 min, 30 min, 1 h, 2 h, and 4 h. After exposure, the supernatant was removed. Cells were gently washed three times with PBS and digested with trypsin-EDTA to obtain a cell suspension. The fluorescence intensity of cells was determined using a BD flow cytometer via the PE channel to assess the degree of PS-MP internalization. Histopathological evaluation Testicular tissues were fixed in 4% paraformaldehyde at room temperature for 24 h, dehydrated, embedded in paraffin, sectioned into 5-µm-thick slices, and stained with hematoxylin and eosin (H&E) for histopathological analyses. To investigate the influence of PS-MPs on the ultrastructure of mouse testes, samples were routinely fixed with 3% glutaraldehyde, followed by 1% osmium tetroxide. After dehydration, ultrastructural changes were analyzed using a transmission electron microscope (TEM). For immunohistochemical (IHC) staining, sections were blocked to reduce non-specific binding and incubated with the primary antibody at 4 °C overnight. After washing, sections were incubated with the HRP-conjugated secondary antibody for 20 min, washed with PBS, developed with DAB, counterstained with hematoxylin, and imaged under a light microscope. For immunofluorescent (IF) staining, sections were incubated with the fluorescent-labeled secondary antibody at room temperature for 1 h, followed by nuclei staining with DAPI. Fluorescent images were captured using a confocal laser scanning microscope. Anti-MAP3K1 antibody was purchased from Santa Cruz Biotechnology (USA). Anti-LC3β antibody was purchased from Proteintech (China). Anti-Cleaved-Caspase3 antibody was purchased from Affinity (China). Anti-Beclin1 antibody was purchased from Wanleibio (China). Cell viability assay GC-1 cells were seeded in 96-well plates (1 × 10^4 cells/well) and treated with different concentrations of 50 nm PS-MPs in serum-free DMEM. After 24 h of incubation, the supernatant was removed. MTT solution was added and the cells were incubated for 4 h. Subsequently, DMSO was added, and the plates were incubated for 10 min. Finally, cell viability was determined by measuring the absorbance at 490 nm using a microplate reader. In addition, after PS-MP treatment, cells were harvested and incubated with the fluorescent probe 7-aminoactinomycin D (7-AAD; Beyotime Biotech., China) to analyze the cell death percentage by flow cytometry (FCM). Apoptosis assay Apoptosis in GC-1 cells was induced by treatment with different concentrations of 50 nm PS-MPs (0, 1, 5, and 25 µg/mL) for 24 h. After incubation, cells were harvested, washed three times with cold PBS, and stained using an Annexin V-FITC Apoptosis Detection kit (Yeasen, Shanghai, China). Cells were incubated in the dark at room temperature for 15 min. Apoptotic cells were then quantified using a BD flow cytometer. Transcriptomics A total of 20 mouse testicular tissue samples were used to construct the RNA-seq library. Transcriptome sequencing was performed on an Illumina high-throughput sequencing platform. Detailed operation procedures can be found in our previous research [[51]5]. Quantitative real-time PCR (qRT-PCR) qRT-PCR was applied to analyze mRNA levels of target genes in mouse testes. Testicular total RNA was extracted using TRIzol reagent and reverse-transcribed into cDNA using a reverse transcription kit (ThermoFisher Scientific, USA). qRT-PCR was performed to quantify mRNA levels using SYBR^® Premix Ex Taq kits (Invitrogen, USA), and the 2^−ΔΔCt method was used for quantifying relative expression. All primers were prepared by Sangon Biotech, China (Supplementary Table 1). Western blot Protein samples were extracted from mouse testes or GC-1 cells using RIPA lysis buffer. Protein concentrations were measured using BCA protein assay kits. Equal amounts of protein (25 µg) from each sample were separated by SDS-PAGE and then transferred to nitrocellulose membranes. The membranes were blocked with 5% non-fat milk at room temperature for 1 h and incubated with primary antibodies at 4 °C overnight. After washing three times with TBST under gentle agitation, the membranes were incubated with HRP-conjugated secondary antibodies for 1 h. Protein bands were visualized using an enhanced chemiluminescence (ECL) reagent with the GeneGnome Bio Imaging System (Syngene, UK). MAP3K1 and FOXA1 antibodies were purchased from Santa Cruz Biotechnology, USA. α-Tubulin, LC3β, Fas, and FasL antibodies were purchased from Proteintech, China. Cleaved-Caspase3 and cleaved-Caspase9 antibodies were purchased from Affinity, China. Beclin1, Bcl-2, p-p38, P62, ATG7, and ATG12 antibodies were purchased from Wanleibio, China. Bax, Bad, Bcl-xL, p38, NUR77, p-c-fos, c-fos, p-c-jun, c-jun, JNK, p-JNK, ERK, and p-ERK antibodies were purchased from Cell Signaling Technology, USA. Co-immunoprecipitation (Co-IP) Proteins were extracted from GC-1 cells and mixed with Protein A agarose beads to preclear nonspecific proteins. After centrifugation, the supernatant was collected and incubated overnight at 4 °C with FOXA1 or MAP3K1 antibodies. Rabbit IgG was used as a negative control. Protein A agarose beads were then added to capture the antigen-antibody complex, followed by incubation at 4 °C under gentle agitation for 2–4 h. The bead-antigen-antibody complex was pelleted by brief centrifugation, washed three times with RIPA buffer, resuspended in loading buffer, mixed thoroughly, and denatured by boiling. Finally, MAP3K1 and FOXA1 were analyzed by Western blot. Chromatin Immunoprecipitation (ChIP) Potential binding sites of NUR77, FOXA1, and MAP3K1 were predicted, and specific primers were designed to target the promoter regions of MAR3K1. GC-1 cells were treated with formaldehyde for 10 min to cross-link DNA and associated proteins, followed by adding glycine to quench residual formaldehyde. Cells were lysed with SDS buffer containing protease inhibitor cocktail II, and the DNA was then fragmented by sonication. The chromatin solution was precleared with protein G agarose beads, and then incubated at 4℃ overnight with antibodies against NUR77 or FOXA1. Normal rabbit IgG was used as a negative control. Subsequently, protein G agarose beads were added and incubated at 4℃ for 2 h with gentle rotation to capture the antibody-protein-DNA complexes. The immunoprecipitates were sequentially washed, eluted, and reverse cross-linked to release the DNA. The purified DNA was amplified by qPCR, and PCR products were analyzed by agarose gel electrophoresis. Primers targeting the proximal promoter regions are listed in Supplementary Table 1. TUNEL assay After deparaffinization, paraffin sections of mouse testicular tissues were incubated with proteinase K at 37 °C for 22 min. Sections were permeabilized with 0.1% Triton X-100 (in PBS) at room temperature for 20 min. After permeabilization, a TUNEL reaction mixture (TdT enzyme: dUTP: buffer = 1:5:50) was added to the sections, which were then incubated at 37 °C for 1 h. Nuclei were counterstained with DAPI. After mounting with antifade mounting medium, the sections were imaged using a fluorescence microscope. DAPI staining yielded blue nuclei under UV excitation, whereas TUNEL staining produced red-positive apoptotic nuclei. Statistical analysis Descriptive statistics were performed to analyze the characteristics of study participants, the distribution of MPs (PS, PVC, PE, PA, PET, PP, and total MPs), and semen quality parameters. PT use frequency (Never, Occasionally, Frequently, and Daily) and levels of PS, PVC, and total MPs (0, 1, 2 particles) were treated as ordinal variables, while conventional semen quality parameters were included as continuous variables. A multivariable linear regression model was applied to assess the associations between PS, PVC, and total MP levels and PT use frequency and semen quality parameters. Due to the non-normal distribution of semen quality data, log-transformation was applied prior to regression analysis. Additionally, given the potential confounding effects of age and BMI on male semen quality and the impact of PT use on MP levels in semen, stratified analyses were conducted based on age (≤ 24 years vs. >24 years), BMI (< 24 kg/m² vs. ≥ 24 kg/m²), and PT use (yes vs. no). Data from animal and cell studies were expressed as mean ± standard deviation (SD). Statistical differences were analyzed by one-way ANOVA using SPSS 23.0. P < 0.05 was considered statistically significant. Results Characteristics of the study population The mean age and BMI of the 200 male participants were 24.59 ± 4.49 years and 22.56 ± 2.97 kg/m², respectively. Over three-quarters of the participants (77%) reported an abstinence period of 3 to 5 days prior to semen collection. The majority of participants were Han Chinese (91.5%). Among the study population, 125 participants (62.5%) reported that they never smoked, and 48 participants (24%) reported that they never drank alcohol. A total of 179 participants (89.5%) had at least a high school education, and 102 participants (51%) reported a monthly household income of over 4,000 yuan (Table [52]1). Additionally, 45 participants (22.5%) had used plastic tableware (PT) within the past three weeks. The median values for sperm concentration, total sperm count, progressive motility, total motility, and immobility were 72.0 × 10⁶/mL, 272.6 × 10⁶/ejaculate, 49.5%, 57.0%, and 43.0%, respectively (Table [53]2). Table 1. Demographic characteristics of study participants (n = 200) Characteristics Mean ± SD or N (%) Age, years 24.59 ± 4.49 ≤ 24 123 (61.5) > 24 77 (38.5) BMI, kg/m^2 22.56 ± 2.97 < 24 147 (73.5) ≥ 24 53 (26.5) Abstinence time 3–5 154 (77) > 5 46 (23) Ethnicity Han 183 (91.5) Others 17 (8.5) Smoking status Never 125 (62.5) Former 34 (17) Current 41 (20.5) Alcohol use Never 48 (24) < 3 times per week 150 (75) ≥ 3 times per week 2 (1) Education level ≤ High school 21 (10.5) >High school 179 (89.5) Household income, RMB Yuan/month < 4000 98 (49) ≥ 4000 102 (51) Ever fathered a child Yes 33 (16.5) No 167 (83.5) Plasticware usage in the past 3 months Yes 45 (22.5) No 155 (77.5) House decoration within the past 3 months Yes 14 (7) No 186 (93) [54]Open in a new tab SD: Standard deviation; BMI: Body mass index Table 2. Distribution of sperm parameters (n = 200) Sperm quality parameters Median 25th 75th Sperm concentration (10^^6 per mL) 72 50 96.5 Sperm count (10^^6 per ejaculate) 272.6 173.3 403.4 Sperm progressive motility (PR, %) 49.5 39 60.25 Sperm total motility (%) 57 42 64 Sperm immotility (%) 43 36 58 [55]Open in a new tab Identification of MPs in semen As shown in Fig. [56]1A, infrared spectroscopy and SEM confirmed the presence of multiple microplastic polymers, including PS, PVC, and PE, in semen samples. Among 200 semen samples, the overall detection rate of MPs was 55.5% (111/200). A total of 128 MPs were identified in semen samples, comprising six polymer types. PS (32.03%) and PVC (36.72%) were the predominant polymers, each accounting for over 30% of the total MPs, followed by PE (23.44%), PET (5.47%), and PP (2.34%). No PA was detected in semen samples (Fig. [57]1B). These findings indicate that human semen is contaminated by MPs, predominantly composed of PS and PVC. Based on these results, subsequent analyses were therefore focused on PS, PVC, and total MPs. PT use frequency was stratified into four categories: Never, Occasionally, Frequently, and Daily. As shown in Fig. [58]1C, semen contents of PS, PVC, and total MPs increased in tandem with PT use frequency, with total MPs significantly higher in daily PT users (P < 0.05). These results highlight a significant correlation between MP accumulation in semen and PT use frequency. Fig. 1. [59]Fig. 1 [60]Open in a new tab Associations between MPs and semen quality among participants. (A) SEM photomicrographs and micro-IR of microplastics in human seminal fluid. (B) Analysis of microplastic content in semen samples. n = 200. (C) ORs (95% CI) for PS, PVC, or total MPs associated with PT use frequency. n = 200. (D) Regression coefficients (95% CI) for sperm motion parameters associated with PS, PVC, or total MPs. n = 200. PS, polystyrene; PVC, polyvinyl chloride; PE, polyethylene; PA, polyamide; PET, polyethylene terephthalate; PP, polypropylene; PT, plastic tableware; OR, odds ratio; CI, confidence interval. The model was adjusted for age, BMI, abstinence duration, alcohol use, smoking status, education level, and reproductive history Analysis of associations between MPs exposure and semen quality parameters After adjusting for covariates (age, BMI, PT usage, income, abstinence period, alcohol consumption, smoking, education level, and reproductive history), multivariable linear regression analyses revealed no significant association between concentrations of PS, PVC, and total MPs in semen and declines in conventional semen quality parameters (Fig. [61]1D). However, stratified analysis revealed that in men aged ≤ 24 years, higher PVC concentrations trended toward compensatory increases in sperm progressive motility (P = 0.04; 95% CI: 0.44, 21.05) and total motility (P = 0.06; 95% CI: −0.44, 19.58). Among participants with a BMI < 24 kg/m², higher total MPs exposure was marginally negatively associated with sperm concentration (P = 0.08; 95% CI: −30.07, 2.4) and positively associated with sperm progressive motility (P = 0.03; 95% CI: 0.77, 16.61) and total motility (P = 0.04; 95% CI: 0.50, 15.67). Furthermore, in participants who used PT, higher total MPs exposure trended toward a decrease in sperm concentration (P = 0.07; 95% CI: −22.65, 1.3), alongside a potential compensatory increase in sperm progressive motility (P = 0.09; 95% CI: −0.99, 14.43) and total motility (P = 0.07; 95% CI: −0.71, 14.01). Accumulation of PS-MPs in mouse testes and their internalization into spermatogonia SEM analyses confirmed that the PS-MPs of three different sizes (50 nm, 5 μm, and 50 μm) exhibited spherical morphology with a uniform size distribution in each size group (Fig. [62]2A). Ex vivo fluorescence imaging revealed that 50 nm, 5 μm, and 50 μm PS-MPs accumulated in mouse testicular tissues (Fig. [63]2B). Moreover, FCM analyses demonstrated that 50 nm PS-MPs were internalized into the cytoplasm of GC-1 cells more rapidly and efficiently than 5 μm and 50 μm PS-MPs, with the highest level of uptake being observed at 30 min post-treatment (Fig. [64]2C). 7-AAD staining and FCM results indicated that the cell death rate of GC-1 cells increased in a concentration-dependent manner after treatment with 50 nm PS-MPs (Fig. [65]2D). Additionally, the MTT assay revealed that the cell viability of GC-1 cells decreased after exposure to 50 nm PS-MPs when compared with the control group (Fig. [66]2E). Fig. 2. [67]Fig. 2 [68]Open in a new tab PS-MPs accumulated in mouse testes and were internalized in the cytoplasm. (A) SEM photomicrographs of PS-MPs. (B) PS-MP accumulation analysis through ex vivo fluorescence images. (C) PS-MP internalization analysis via FCM. n = 3/group. (D) Effects of PS-MPs on cell death by 7-AAD. n = 3/group. (E) Effects of PS-MPs on cell death and cell viability of GC-1 cells. n = 6/group. Error bars represented the standard deviation. ^*p < 0.05 compared with the control Effects of PS-MPs on sperm quality and testicular histomorphology Exposure to PS-MPs significantly reduced sperm quality in mice. As shown in Fig. [69]3A, 50 nm PS-MPs exposure decreased sperm concentration, total motility, and progressive motility by 33.23%, 21.49%, and 37.82%, respectively, compared with the control group. The sperm malformation rate in the 50 nm group was also significantly higher than in the control group. Light microscopy revealed normal sperm morphology in the control, characterized by hook-shaped heads and smoothly curved tails (red arrows). In contrast, sperm from PS-MPs-exposed mice exhibited various abnormalities, such as folding, dual tails, and amorphous shapes (black arrows) (Fig. [70]3B). Fig. 3. [71]Fig. 3 [72]Open in a new tab Effects of PS-MPs on sperm quality and testicular histomorphology. (A) Effects of PS-MPs on mouse sperm concentration, total motility, progressive motility, and sperm malformation rate. n = 12/group. (B) Effects of PS-MPs on mouse sperm morphology. The red arrow indicated normal sperm, and the black arrow indicated abnormal sperm. n = 12/group. (C) Histological changes in mouse testes were assessed using HE staining (magnification: × 400). (D) TEM analysis in mouse testes (magnification: × 1500, × 5000). Error bars represented the standard deviation. ^*p < 0.05 compared with the control PS-MPs exposure also induced abnormal testicular histology and ultrastructure in mice. HE staining analysis showed that exposure to PS-MPs (particularly 50 nm particles) caused atrophy and deformation of seminiferous tubules, disorganized arrangement of spermatogenic cells, and impaired spermatogenesis (Fig. [73]3C). TEM analysis further revealed irregular nuclear shapes and partial nuclear membrane rupture in PS-MPs-exposed testicular tissues. Additionally, increased autolysosomes (red arrows) were observed in both the 50 nm and 5 μm groups (Fig. [74]3D). These results suggest that PS-MPs may promote autophagy in mouse testicular tissues. PS-MPs induced autophagy in mouse testes and spermatogonia Transcriptomic analyses were performed on mouse testicular tissues using RNA-seq. As shown in Supplementary Fig. 1, exposure to 50 nm, 5 μm, and 50 μm PS-MPs identified 985, 1299, and 929 differentially expressed genes (DEGs), respectively, compared with the control group (P < 0.05). To investigate associated metabolic and signaling pathways, KEGG pathway enrichment analyses were conducted. Intriguingly, DEGs in the 50 nm group were predominantly enriched in apoptosis-related signaling pathways compared with those in the control group (Fig. [75]4A). A heatmap focusing on autophagy- and apoptosis-related genes further showed their differential expression in the 50 nm group (Fig. [76]4B). The findings suggest that autophagy and apoptosis may play a potential role in the sperm quality reduction mediated by 50 nm PS-MPs. In addition, to explore upstream signaling events, we examined DEGs related to the MAPK/MAP3K1-p38 pathway. Although FOXA1, MAP3K1, and p38 themselves were not identified as DEGs, several MAPK-related genes, such as Map3k14, Dusp4, Foxj1, showed significant alterations in the 50 nm group (Supplementary Fig. 2). Fig. 4. [77]Fig. 4 [78]Open in a new tab PS-MPs induced apoptosis and autophagy in mouse testes and spermatogonia. (A) KEGG pathway enrichment analysis and functional analysis. The red indicated apoptosis. n = 5/group. (B) Heatmap of apoptosis- and autophagy-related genes with different expressions after PS-MPs exposure. n = 5/group. (C) Effects of PS-MPs on mRNA levels of ATG5, ATG7, and Beclin1 in mouse testes. n = 12/group. (D) Effects of PS-MPs on the protein expression of LC3β and P62 in mouse testes. n = 6/group. (E) IF analysis of LC3β in mouse testes. The green indicated LC3β positivity. The blue represented nuclei. (Magnification: × 400). (F) IHC analysis of Beclin1 in mouse testes. (Magnification: × 400). (G) Effects of PS-MPs on autophagy-related biomarkers in spermatogonia. n = 3/group. Error bars represented the standard deviation. ^*p < 0.05, compared with the control Given the above findings that 50 nm PS-MPs exposure led to decreased sperm quality and that autophagy- and apoptosis-related genes were significantly enriched, molecular mechanisms underlying 50 nm PS-MP effects were further explored. Among the autophagy-related genes, ATG5, ATG7, and Beclin1 were identified as DEGs in response to PS-MPs, while LC3β and P62 were not identified as DEGs. These two molecules were nevertheless selected for validation because they are well-established canonical markers of autophagosome formation (LC3β) and autophagic flux/degradation (P62). Thus, our validation strategy combined transcriptome-driven targets with functionally representative autophagy regulators. In mouse testicular tissues, mRNA levels of autophagy-related genes ATG5 and ATG7 were significantly upregulated in all PS-MP groups. Beclin1 mRNA levels were markedly increased in the 50 nm and 50 μm groups (Fig. [79]4C). At the protein level, the expression of autophagy-related LC3β and P62 in the 50 nm group was increased by 69.49% and 138.28%, respectively, compared with the control group (Fig. [80]4D). IF analyses further confirmed that PS-MPs, particularly 50 nm PS-MPs, elevated LC3β expression in mouse testes (Fig. [81]4E). Additionally, IHC analyses revealed that the brown-positive staining of Beclin1 in mouse testes was enhanced following PS-MPs exposure (Fig. [82]4F). Consistently, upregulation of LC3β, P62, and ATG7 protein expression was observed in GC-1 cells treated with 50 nm PS-MPs, while no significant change in ATG12 was observed (Fig. [83]4G). PS-MPs induced apoptosis in mouse testes and spermatogonia TUNEL staining revealed a significant increase in the levels of apoptotic markers in the testicular tissues of PS-MPs-exposed mice compared with the control (Fig. [84]5A). Annexin V/PI staining further demonstrated a concentration-dependent increase in apoptotic GC-1 cells following 50 nm PS-MP treatment (Fig. [85]5B). Analyses of apoptotic proteins in testicular tissues showed that in the 50 nm PS-MP group, pro-apoptotic proteins Bad and Bax were upregulated, while anti-apoptotic proteins Bcl-2 and Bcl-xL were downregulated, compared with the control (Fig. [86]5C). Generally, apoptosis is mediated by both intrinsic mitochondrial pathways and extrinsic death receptor pathways. Specifically, the protein expression of intrinsic apoptotic markers cleaved-Caspase3 and cleaved-Caspase9 was increased by 85.43% and 72.73%, respectively, in the 50 nm PS-MP group compared with the control, whereas no significant alterations in extrinsic apoptotic proteins Fas and FasL were observed following PS-MPs exposure (Fig. [87]5D). IF analyses of mouse testes also confirmed the upregulation of cleaved-Caspase3 induced by PS-MPs, especially in the 50 nm group (Fig. [88]5E). Consistently, in GC-1 cells treated with 50 nm PS-MPs, the protein expression of Bad, Bax, cleaved-Caspase3, and cleaved-Caspase9 was elevated, while Bcl-2 and Bcl-xL were decreased (Fig. [89]5F and G). Fig. 5. [90]Fig. 5 [91]Open in a new tab PS-MPs induced apoptosis in mouse testes and spermatogonia. (A) Testicular TUNEL staining. (Magnification: × 400). (B) Apoptosis analysis of GC-1 cells by FCM. n = 3/group. (C-D) Effects of PS-MPs on apoptosis-related biomarkers in mouse testes. n = 6/group. (E) IF analysis for cleaved-Casepase3 in mouse testes. The red indicated cleaved-Casepase3 positivity. The blue represented nuclei. (Magnification: × 400). (F-G) Effects of PS-MPs on apoptosis-related biomarkers in spermatogonia. n = 3/group. Error bars represented the standard deviation. ^*p < 0.05; compared with the control PS-MPs activated the MAP3K1/p38/c-fos signaling pathway In vivo and in vitro analyses showed that PS-MPs activated the MAP3K1/p38/c-fos signaling pathway. In mouse testes, the protein expression of MAP3K1 and p-p38 was markedly elevated in the 50 nm PS-MP group compared with the control group (Fig. [92]6A). In contrast, PS-MPs showed minimal effect on the protein expression of p-ERK and p-JNK in mouse testes (Supplementary Fig. 3). Similarly, in GC-1 cells treated with 50 nm PS-MPs (25 µg/mL), the protein expression of MAP3K1 and p-p38 was increased by 1.22- and 0.91-fold, respectively, relative to the control (Fig. [93]6B). IHC analyses confirmed MAP3K1 upregulation in PS-MPs-exposed mouse testes (Fig. [94]6C). Moreover, in mouse testes exposed to 50 nm PS-MPs, there was a significant increase in the protein expression of p-c-fos, while p-c-jun expression remained largely unchanged (Fig. [95]6D). This pattern was also observed in PS-MPs-treated spermatogonia. PS-MPs significantly upregulated p-c-fos expression without notably affecting p-c-jun in GC-1 cells (Fig. [96]6E). To further investigate the connection between p38 and c-fos, GC-1 cells were pretreated with the p38 inhibitor adezmapimod, which effectively inhibited the PS-MPs-induced p38 phosphorylation (Fig. [97]6F, Supplementary Fig. 4). Moreover, p-c-fos expression was significantly reduced, whereas p-c-jun was not significantly affected (Fig. [98]6G). These results indicate that PS-MPs activate the MAP3K1/p38 pathway, leading to c-fos phosphorylation and subsequent transcriptional activation. Fig. 6. [99]Fig. 6 [100]Open in a new tab Effects of PS-MPs on the MAP3K1/p38/c-fos pathway in mouse testes and spermatogonia. (A-B) Effects of PS-MPs on the protein expression of MAP3K1 and p38 in mouse testes (n = 6/group) and spermatogonia (n = 3/group). (C) IHC analysis of MAP3K1 in mouse testes. (Magnification: × 400). (D-E) Effects of PS-MPs on the protein expression of c-fos and c-jun in mouse testes (n = 6/group) and spermatogonia (n = 3/group). (F-G) Analysis of regulatory effects of p38 on c-fos and c-jun using a p38 inhibitor (Adezmapimod, 1 µM). n = 3/group. Error bars represented the standard deviation. ^*p < 0.05; compared with the control PS-MPs induced the MAP3K1/p38/c-fos pathway via FOXA1 As illustrated in Fig. [101]7A, the protein expression of the transcription factor FOXA1 in mouse testes was increased by 58.74% and 62.87% in the 50 nm and 50 μm PS-MP groups, respectively, compared with the control group. A concentration-dependent upregulation of FOXA1 protein expression was also observed in GC-1 cells treated with 50 nm PS-MPs (Fig. [102]7B). Conversely, no significant changes were detected in the expression of the transcription factor NUR77 in mouse testes and spermatogonia following PS-MPs exposure. As shown in Fig. [103]7C, the GRAMM (Global Range Molecular Modeling) protein docking analyses indicated that MAP3K1 could specifically interact with either FOXA1 or NUR77. To elucidate the regulatory relationship between MAP3K1 and these transcription factors, ChIP and qPCR assays were conducted in GC-1 cells. The findings revealed that FOXA1, but not NUR77, directly bound to the MAP3K1 promoter, thus modulating MAP3K1 transcription (Fig. [104]7D). Further bidirectional Co-IP assays in GC-1 cells, using anti-MAP3K1 and anti-FOXA1 antibodies, confirmed that MAP3K1 physically interacted with FOXA1 (Fig. [105]7E). To explore the role of FOXA1 in the MAP3K1/p38/c-fos pathway, three FOXA1-targeting siRNAs were synthesized, with the most effective (siRNA3) selected (Supplementary Fig. 5). Co-treatment with FOXA1 siRNA3 inhibited the activation of the downstream MAP3K1/p38/c-fos pathway by PS-MPs, resulting in a significant reduction in the protein expression of MAP3K1, p-p38, and p-c-fos in GC-1 cells, while p-c-jun expression remained unchanged (Fig. [106]7F and G). Fig. 7. [107]Fig. 7 [108]Open in a new tab PS-MPs induced the MAP3K1/p38/c-fos pathway via the transcription factor FOXA1. (A-B) Effects of PS-MPs on the protein expression of FOXA1 and NUR77 in mouse testes (n = 6/group) and spermatogonia (n = 3/group). (C) GRAMM docking analysis of MAP3K1 and FOXA1 or NUR77. (D) Interactions of FOXA1 and NUR77 with the MAP3K1 promoter. (E) Bidirectional Co-IP analysis of MAP3K1 and FOXA1. (F-G) Effects of the FOXA1 siRNA3 on FOXA1, MAP3K1, P38, c-fos, and c-jun in spermatogonia. n = 3/group. Error bars represented the standard deviation. ^*p < 0.05; compared with the control Discussion Exposure to MPs has been identified as a potential risk factor for reduced sperm quality in men. However, epidemiological studies specifically investigating the effect of MPs in semen on sperm quality are still scarce. Furthermore, the mechanisms underlying the reproductive toxicity of MPs remain unclear. To address these gaps, this study first evaluated the impact of MPs on sperm quality through population-based epidemiological analyses, and then further investigated the molecular mechanisms using murine models and spermatogonial cell cultures (Fig. [109]8). Fig. 8. [110]Fig. 8 [111]Open in a new tab A schematic model illuminating the harmful effects of plastic tableware use and MPs on sperm quality and underlying mechanisms In this study, we analyzed the abundance of six common MPs in semen samples from 200 volunteers. The overall detection rate of MPs was 55.5%, which was higher than the 44% detection rate observed in a cohort of 30 volunteers from Peking University [[112]16], but was significantly lower than the 100% detection rate observed in 40 volunteers from Jinan [[113]17]. These discrepancies in detection rates across studies may stem from methodological differences (e.g., infrared [IR] versus Raman spectroscopy), variations in the study populations (e.g., general population versus subfertile men), and geographic factors (e.g., coastal versus inland regions) [[114]18]. Moreover, previous studies were limited by small sample sizes of 30 to 40 participants, which potentially introduced bias and contributed to inconsistent results. Considering the relatively large number of biological samples analyzed in this study, we employed micro-IR combined with SEM to ensure a balance between analytical accuracy, reproducibility, and processing efficiency. However, although micro-IR spectroscopy provided reliable identification of microplastics in this study, this technique had limitations in detecting nanoplastics due to its relatively low spatial resolution. Therefore, the 50 nm PS-MPs used in our animal and cell experiments could not be directly detected in the human semen samples analyzed in the current study. This methodological limitation should be considered when interpreting the translational relevance of our findings. In future work, we will employ complementary techniques (e.g., Raman spectroscopy, pyrolysis-GC/MS) to characterize smaller particles and address potential nanoplastic exposure. In our cohort, PS and PVC microplastics were the most prevalent, each accounting for over 30% of the detected MPs, suggesting their dominance in the studied population’s exposure. PS and PVC are commonly used in daily life: PVC is primarily used in pipes, cable insulation, and flooring, while PS is widely used in disposable tableware, foam packaging, and building materials [[115]19]. Based on these findings, we selected PS and PVC for further mechanistic investigation. With the growing internet penetration and the rapid rise of the stay-at-home economy, food delivery services have become an increasingly significant component of consumer behaviors. According to estimates based on the average microplastic content in food delivery containers and ordering frequency among high-risk individuals (4 to 7 times per week), each person could ingest between 12 and 203 MPs weekly from these containers, with PS (polystyrene) containers accounting for the highest proportion [[116]7]. A cohort study reported that alterations in gut and oral microbiota were closely associated with PT use frequency, potentially leading to gastrointestinal dysfunction and respiratory symptoms [[117]8]. A recent simulation study demonstrated that hot water in disposable paper cups could leach MPs into beverages [[118]6]. When orally administered to pregnant mice, these MPs exhibited tissue-specific accumulation, posing risks to fetal development [[119]20]. Our study further revealed a strong correlation between PT use frequency and MP accumulation in semen samples. Although previous epidemiological studies detected MPs in semen samples from both the general population and men undergoing assisted reproductive technologies (ART), they failed to establish clear associations between sperm quality and specific MP types. In our cohort, frequent PT users showed significant negative correlations between total MPs exposure and sperm concentration, suggesting that elevated MP accumulation from frequent PT use could reduce sperm counts. Intriguingly, we observed positive correlations between total MPs exposure and both sperm progressive motility and total motility in PT users with BMI < 24 kg/m². Three potential mechanisms may account for this phenomenon: first, selective survival, where only the higher quality, more motile sperm survive when sperm count is low; second, the influence of testosterone regulation on sperm motility, where changes in testosterone may enhance sperm motility even when sperm count decreases; and third, a compensatory mechanism, where exposure to environmental pollutants may reduce sperm count through testicular inflammation, but the remaining sperm may increase their energy metabolism to boost motility. Given that PS-MPs were the most prevalent MP type detected in human semen in this study and considering the widespread use of PS in plastic tableware, we subsequently investigated PS-MPs’ reproductive toxicity and its underlying molecular mechanisms using animal and cell models. In rodent models, PS-MPs have been demonstrated to exert male reproductive toxicity, as evidenced by structural alterations in the seminiferous tubules, impaired Sertoli cell function, reduced testosterone levels, increased sperm apoptosis, DNA damage, and decreased sperm count. Aligning with these findings, our in vivo study also revealed that exposure to PS-MPs resulted in diminished sperm motility, lowered sperm concentration, and elevated sperm malformation rate. The p38 signaling pathway, a multifunctional cascade, regulates a variety of biological processes, including stress responses, inflammation, apoptosis, and autophagy. Within the reproductive system, the p38 pathway maintains the stability and balance of spermatogenesis by balancing apoptosis and autophagy. For instance, L-theanine and dihydromyricetin played a protective role against heat stress-induced reproductive dysfunction through the p38/NF-κB/Nrf2/HO-1 pathway, which mediated the Caspase3 apoptotic pathway of germ cells [[120]21]. Additionally, p38 modulates the formation and maturation of autophagosomes by regulating ATG (autophagy-related genes) proteins. During spermatogenesis, proper autophagy facilitates the removal of damaged mitochondria and other organelles [[121]9]. However, excessive activation of the p38 pathway induces overactive apoptosis and autophagy in germ cells, ultimately leading to cell death and compromised sperm quality and male fertility. In our study, although exposure to 50 nm PS-MPs did not significantly affect levels of reproductive hormones (GnRH, FSH, LH, and testosterone) and levels of oxidative stress markers (GSH-PX, CAT, and SOD) [[122]5], it did activate the p38 pathway and did promote autophagy and apoptosis in mouse testes and spermatogonia. Notably, these effects were attenuated by the p38 inhibitor Adezmapimod, suggesting that PS-MPs enhance autophagy and apoptosis through the p38 pathway. To validate these findings, we selected ATG5, ATG7, and Beclin1, which were enriched as DEGs in our transcriptomic analysis, together with LC3β and P62, the canonical autophagy regulators, thereby combining transcriptome-driven targets with functionally representative markers. MAP3K1 acts as an upstream kinase in the three-tiered MAPK signaling cascade. Within the p38 pathway, MAP3K1 activates MAPKKs (such as MKK3 and MKK6) through phosphorylation, which in turn activate p38, thereby regulating cellular processes like apoptosis, autophagy, inflammation, and cell differentiation [[123]22]. In the present study, we observed a significant increase in MAP3K1 protein expression in mouse testicular tissues and spermatogonia following exposure to 50 nm PS-MPs, indicating that PS-MPs may activate the p38 pathway via MAP3K1. Although FOXA1, MAP3K1, and p38 were not identified as DEGs, several MAPK-related genes (e.g., Map3k14, Dusp4, Foxj1) were enriched in the 50 nm group, providing transcriptomic support for our experimental focus on the FOXA1/MAP3K1/p38 axis [[124]23–[125]25]. Specifically, the enrichment of these genes indicates that MAPK-related networks were activated under PS-MPs exposure, further reinforcing this rationale. To further investigate the upstream regulatory mechanisms by which PS-MPs induced MAP3K1 expression, we used the ENCODE database to predict transcription factors potentially binding to the MAP3K1 promoter. GRAMM docking analyses were then employed to simulate the binding of these factors to the promoter. Our predictions indicated that FOXA1 and NUR77 could bind to the promoter region of MAP3K1, both of which were highly expressed in mouse testicular tissues and spermatogonia following PS-MPs exposure. A previous study reported that FOXA1 influenced sperm production efficiency by interacting with estrogen or androgen receptors, thereby regulating hormone-responsive gene expression [[126]26]. Under stress conditions, such as oxidative stress or toxin exposure, NUR77 overexpression in germ cells impaired spermatogenesis and reduced fertility [[127]27]. In our research, ChIP and qPCR assays were conducted to validate the interactions of FOXA1 and NUR77 with the MAP3K1 promoter. The results demonstrated that FOXA1, but not NUR77, directly bound to the MAP3K1 promoter, thus modulating MAP3K1 transcription. Moreover, bidirectional Co-IP assays were further used to explore the protein interaction between FOXA1 and MAP3K1, revealing that FOXA1 physically bound to and interacted with MAP3K1 in spermatogonia. By knocking down FOXA1 expression in PS-MPs-exposed spermatogonia using a specific FOXA1 siRNA, our rescue experiments confirmed that FOXA1 positively regulated the p38/c-fos pathway via MAP3K1. To our knowledge, this is the first study to identify FOXA1 as a critical regulator in the process by which PS-MPs exposure reduces sperm quality. It should be noted that there are several limitations in our research. In our in vivo studies, there was a lack of systematic dose-response analysis. We tested only a single dose of 50 nm PS-MPs, which restricted the extrapolation of our findings for quantitative risk assessment. Future studies should incorporate multiple exposure doses to better define threshold levels and dose-dependent biological effects. In our in vitro studies, we used only GC-1 cells, which do not fully recapitulate the complex, multicellular environment of the testis. In future research, we plan to include additional complementary models, such as TM3 Leydig cells, TM4 Sertoli cells, or advanced 3D co-culture and organoid systems, to better reflect the physiological responses of testicular cells to PS-MPs [[128]28, [129]29]. Moreover, the concentrations of PS-MPs used in in vitro studies were higher than environmental exposure levels. This limits the translational relevance of our findings. In addition, subcytotoxic concentrations were not assessed, which represents a methodological limitation. Future studies should incorporate cytotoxicity screening and environmentally relevant levels (e.g., in the nanogram range) to improve health risk extrapolation. Another limitation is the lack of functional fertility assessments in the present study, such as pregnancy rates or embryo development outcomes, thereby limiting the interpretation of the biological relevance of sperm abnormalities. In future studies, such assessments should be incorporated to enhance translational value. In summary, this study uncovered several critical findings. Our population-based epidemiological research demonstrated a high detection rate of MPs in human semen, with a significant positive correlation between PT use frequency and MP accumulation in semen samples. Further stratified analyses revealed that in individuals with BMI < 24 kg/m² and frequent PT use, total MPs exposure was strongly associated with reduced sperm concentration. In PS-MPs-exposed mice and spermatogonia, we observed a decline in sperm quality, an increased sperm malformation rate, and elevated levels of apoptosis and autophagy, compared with the control group. Mechanistically, 50 nm PS-MPs activated the MAP3K1/p38/c-fos pathway via FOXA1, ultimately promoting apoptosis and autophagy in spermatogonia (Fig. [130]8). By integrating epidemiological evidence with mechanistic insights into the FOXA1/p38 signaling pathway, our work contributes to a better understanding of the detrimental effects of MPs on sperm quality and male reproductive health. Supplementary Information [131]Supplementary Material 1^ (475.3KB, pdf) Acknowledgements