Epidermal growth factor receptor ligands enriched in follicular fluid exosomes promote oncogenesis of fallopian tube epithelial cells.

Based on characteristic driver mutations in HGSC and STIC, we previously established a partially transformed FTE secretory cell line, FE25 (RRID: CVCL_C6JL), in which the p53 and Rb pathways were disrupted by HPV16 E6/E7 [ 10 ]. With chromosomal instability characteristics similar to those observed in HGSC, FE25 cells accumulate DNA copy number variations (CNV) upon increasing number of culture passages [ 12 ] and become fully transformed after 115 passages [ 33 ]. By showing a high response to FF transformation, we classified passage 90–115 FE25 cells as STIC mimic cells in previously published studies [ 10 – 14 , 34 ]. Another FTE cell line used in this study was FT-282-CCNE1 cells, a kind gift from Dr. Ronny Drapkin. This cell line harbors the TP53R175H mutation and overexpresses CCNE1, a downstream effector after Rb [ 33 ]. Cells were characterized without contamination by other cells via short tandem repeat analysis with genotype results compared with those from the ATCC database. Both cell lines were maintained in MCDB105 and M199 media (Sigma) supplemented with 10% fetal bovine serum (FBS), 100 IU/ml penicillin, and 100 µg/mL streptomycin. DAPI staining and fluorescence microscopy were used to ensure that there was no mycoplasma contamination before cell experiments. For inhibition experiments, cells in 1 mL of serum-free medium were pretreated with inhibitors for 30 min. After adding 100 µL of FF or exosomes, the cells were incubated for 30 min at 37 °C. The cells were then collected and analyzed for downstream signaling. MET inhibitor (AMG337), IGF1R inhibitor (PPP), EGFR inhibitor (pelitinib or EKB-569), PDGFR inhibitor (CP-673451), and VEGFR inhibitor (Semaxinib or SU5416) were purchased from MedChemExpress. The drugs were prepared in dimethyl sulfoxide and stored at − 20 °C. Follicular fluid (FF) samples were collected from women undergoing in vitro fertilization (IVF). The samples were obtained during the first fluid aspiration, before oocyte retrieval, with approximately 5 mL of FF collected from each participant for the study. The clinical information of IVF patients, including age, cause of infertility, and ovarian stimulation protocols is presented in Table  1 . Ovarian follicles were aspirated under sonographic guidance after HCG injection. The follicles were individually aspirated by introducing a transvaginal needle. To exclude blood contamination, FF specimens were checked using spectrophotometry by measuring the absorption of Hb at OD 418 nm [ 17 ]. FF from 10 subjects were equally pooled, centrifuged at 1500 g for 15 min to remove cell pellets, filtered using a 0.2-µm size filter, aliquoted, and stored at − 80 °C before use. For experiments, a 100-µL aliquot was stored at 4 °C and used within 1 week to avoid repeated freeze-thaw cycles, following established protocols for consistency [ 13 ]. For the FF exosome experiments, 500 µL of FF was used to isolate exosomes, which were then resuspended in an equal volume of PBS and freshly used. The same volume of 10% FF was used in the transformation activity tests as in our previous experiments [ 12 , 14 ]. The collection and use of specimens were approved by the Institutional Review Board of Tzu Chi Medical Center, Taiwan (approval number: IRB-106‐07‐A, IRB108‐12‐A). Table 1 Clinical information of FF-collected patients ID Age Etiology of infertility a, b Controlled ovarian hyper-stimulation protocol c E2 (ng/mL) 230,018 35 HSG: Both tubes obstructed (history of surgery for ovarian teratoma and chocolate cyst) Elonva(D2) + CC 0.25 (Day8 ~ 12) + Pergoveris 2#(D8 ~ D11)) + Lupron (D12). --> D14 TVOR 1560.5 230,021 40 Patent right fallopian tube; left tube obstructed Elonva(D5) + CC 0.25 (Day 11 ∽ 15) + Pergoveris 2#(D12∽15)) + Lupron (D16). --> D18 TVOR 3548.3 230,022 37 HSG: Both tubes obstructed Elonva(D4) + CC 0.25 (Day8 ∽ 16) + Pergoveris 2#(D8 ∽ D15)) + Lupron (D16). --> D18 TVOR 4185.2 230,024 38 AMH: 1.81ng/mL Elonva(D2) + CC 0.25 (Day9 ∽ 15) + Pergoveris 2#(D9 ∽ 13) + Lupron (D14). --> D15 TVOR 1169.6 230,029 36 HSG revealed patency of fimbrial ends but no spillage Elonva(D4) + CC 0.25 (Day11 ∽ 43) + Pergoveris 2#(D11∽)->3#(D25 ∽ 42) + Lupron (D43). --> D45 TVOR 983 230,035 34 Male factor oligoasthenospermia Lupron 0.4(D5 ∽ D16) + Elonva 150 (D5) + Pergoveris (D9 ∽ 16) + Ovidrel 250 (Day 17) --> D19 TVOR 6083.5 230,037 38 Left ovarian endometrioma, right ovarian cyst, status-post ultrasound-guided aspiration Lupron 0.4(D4 ∽ D27) + Elonva 150 (D4) + Pergoveris (D11∽) 2# -->3# (D24 ∽ 27) + Ovidrel 250 (Day 28) --> D30 TVOR 2962.8 230,038 28 Teratospermia + PCOS Lupron (D14) + Elonva 150 (D3) + Orgalutran 0.25(D9 ∽ D14) + Pergoveris (D10 ∽ D13) 3# --> D16 TVOR 10134.9 230,041 43 Advanced age P1 (IVF-ICSI-PGT pregnancy) Elonva(D5) + Pergoveris (D12 ∽ D16) + CC(D12 ∽ 17) + Lupron (D17) --> TVOR(D19) 2296.7 230,047 45 Advanced age AMH0.618 ng/mL Genotropin (D4 ∽ D12) + Clomifene 2# (D4 ∽ 7) + Elonva 150 (D7) + CC 0.25(Day11) + Lupron (D11) --> TVOR(D13) 619.7 a. No endometriosis cases were included, and no oral contraceptives were administered prior to in vitro fertilization (IVF) cycle; HSG: hysterosalpingogram b. All PCOS (polycystic ovary syndrome) patients were given metformin at a dose of 500 mg thrice a day for at least 1 month before the IVF cycle c. CC: clomiphene citrate; TVOR: ultrasound-guided transvaginal oocyte retrieval Clinical information of FF-collected patients Left ovarian endometrioma, right ovarian cyst, status-post ultrasound-guided aspiration a. No endometriosis cases were included, and no oral contraceptives were administered prior to in vitro fertilization (IVF) cycle; HSG: hysterosalpingogram b. All PCOS (polycystic ovary syndrome) patients were given metformin at a dose of 500 mg thrice a day for at least 1 month before the IVF cycle c. CC: clomiphene citrate; TVOR: ultrasound-guided transvaginal oocyte retrieval For exosome extraction from FF, a 10-kDa filter was first used to exclude small-molecule fractions of FF (FF < 10 kDa). The remaining FF was added to 1/4 volume of SBI ExoQuick precipitation reagent (# EXOQ5A-1, SBI, USA). After centrifugation (1500 g, for 30 min), exosomes were enriched in the pellet (Exo-1). The exosome-poor supernatant (Exo-P) was further treated with a protein G-conjugated CD9 antibody to pull down the residual exosomes. After centrifugation, the supernatant was designated exosome-depleted (Exo-D), and the pellet was designated exosome 2 (Exo-2) (Fig.  1 A). Fig. 1 FF exerts AIG transformation activity on FTE cells through three RTK signals. FF-induced anchorage-independent growth (AIG) of fallopian tube epithelium (FTE) cells was inhibited by some class-specific receptor tyrosine kinase (RTK) inhibitors. ( A , C ) Images show AIG colonies of FE25 cells and FT282-CCNE1 cells treated with FF and class-specific RTK inhibitors. Scale bar 100 μm. ( B , D ) The panel of reduction effects in FF-induced AIG was achieved using different RTK inhibitors. RTK inhibitors AMG337, PPP, EGFRi, PDGFRi, and VEGFRi were used. (EGFR inhibitor EKB-569, 2.5 µM, RTK class I) (MET inhibitor AMG337,10 µM, RTK class VIII) (IGF-1R inhibitor PPP, 100 nM, RTK class II) (PDGFR inhibitor CP-673451, 4.5 µM, RTK class III) (VEGFR inhibitor SU5416, 2.5 µM, RTK class IV) * p  < 0.05, ** p  < 0.01, *** p  < 0.001, comparison as shown in the Figure FF exerts AIG transformation activity on FTE cells through three RTK signals. FF-induced anchorage-independent growth (AIG) of fallopian tube epithelium (FTE) cells was inhibited by some class-specific receptor tyrosine kinase (RTK) inhibitors. ( A , C ) Images show AIG colonies of FE25 cells and FT282-CCNE1 cells treated with FF and class-specific RTK inhibitors. Scale bar 100 μm. ( B , D ) The panel of reduction effects in FF-induced AIG was achieved using different RTK inhibitors. RTK inhibitors AMG337, PPP, EGFRi, PDGFRi, and VEGFRi were used. (EGFR inhibitor EKB-569, 2.5 µM, RTK class I) (MET inhibitor AMG337,10 µM, RTK class VIII) (IGF-1R inhibitor PPP, 100 nM, RTK class II) (PDGFR inhibitor CP-673451, 4.5 µM, RTK class III) (VEGFR inhibitor SU5416, 2.5 µM, RTK class IV) * p  < 0.05, ** p  < 0.01, *** p  < 0.001, comparison as shown in the Figure The exosome morphology was confirmed by transmission electron microscope (TEM): 1–2 µL of exosomes in PBS was placed onto a carbon-coated copper grid, and 5 µL of 2% phosphotungstic acid was added to the grid for 2–3 min and dried at room temperature. The copper grid was observed under TEM operated at 80 kV using a Hitachi S-4700 SEM (Hitachi, Tokyo, Japan). The exosomes were subjected to nanoparticle tracking analysis (NTA) to determine their size and concentration. A NanoSight NS300 instrument (Marvell Panalytical, Malvern) with a 488 nm laser and an sCMOS camera module (Malvern Panalytical) was used according to the manufacturer’s instructions. Three 60-second videos were recorded for each sample at camera level of 14–15. Data analysis was performed using the NTA 3.2 analytical software (Malvern) with a detection threshold of 3. For a exosome-free control in the experiment, the Exo-P was further removed from its remaining exosomes by immunoprecipitation to generate exosome-free FF (Exo-D); Protein G beads (Santa Cruz, sc-2002) conjugated exosome-specific CD9 antibody (2 µg) was added into 1 mL of Exo-P. An equal volume of 10% FF and exosome fractions was used for the transformation activity tests, and an equal amount of protein was used for western blot analysis. Protein concentrations were quantified using Bradford assay dye (BIO-RAD,500-006). Finally, freshly prepared exosomes were used in the experiments. Cells were cultured in a six-well plate with or without treatment. The supernatant of cell lysates treated with RIPA lysis solution was analyzed for protein concentration using Bradford assay dye (BIO-RAD,500-006). An equal volume of 2× Laemmli sample buffer was added to the lysate samples, heated at 95 °C for 5 min, and cooled on ice for 15 min. Crude protein extract (30 µg) was separated by SDS-PAGE and transferred onto a polyvinylidene fluoride (PVDF) membrane. After blocking with 5% skim milk powder dissolved in Tris-buffered saline with 0.05% tween (TBST) for 1 h, the PVDF membrane was incubated with the primary antibody at 4 °C overnight. The antibodies in this study included p-AKT (sc‐514032, Santa Cruz), p‐STAT3 (#9145, Cell Signaling), p‐mitogen‐activated protein kinase (MAPK) (ab223500, Abcam), p‐FAK (GTX100764, Gene Tex), CD-81 (GTX637265, Gene Tex), CD-9 (A19655, Abclonal), HSP70 (GTX637059, Gene Tex), EGFR (A11352, Abclonal), pEGFR (AP0026, Abclonal), Versican (A19655, Abclonal), Amphiregulin (A1860, Abclonal), BTC (A2588, Abclonal), EREG (A16372, Abclonal), TGFA (A0337, Abclonal), and actin (#4970, Cell Signaling). The membrane was washed with TBST, exposed to appropriate horseradish peroxidase (HRP)‐conjugated secondary antibodies, and stained with the ECL Western blot detection reagent (GE Healthcare, RPN2209). First, 1 ⋅ 10 6 FE25 cells were cultured in a 6-well plate. FF exosomes isolated from 500 µL of FF were stained with 1 µM calcein AM (Thermo Fisher Scientific) in 500 µL PBS at room temperature for 1 h, following a modified method [ 35 ]. After washing, the EVs were resuspended in the same volume of PBS and added to the cells at 10% of the well volume. Afterward, the cells were incubated at 37 °C for 1 h, washed with PBS, and fresh medium was added. Exosome uptake was then observed using an inverted fluorescence microscope at 400⋅ magnification with differential interference contrast (DIC) and EGFP channels. Culture-Insert 2 wells (Ibidi, Cell in focus) were used to perform a cell migration assay. A total of 15,000 cells in 60 µL of serum-free medium were added to each side of the culture-insert well, with a waiting period of 3–4 h to achieve complete cell adhesion. After removal of the inserts, cells were pretreated with the inhibitor for 30 min in serum-free medium, and the tested reagent was added. The gap was closed after 24 h. A clonogenic assay was used to analyze the ability of single cells to grow as colonies in a 2D attachment culture. Five hundred cells in serum-free medium were seeded in a six-well plate with or without the tested reagent. For the inhibition assay, cells were pretreated with the inhibitor for 30 min. The culture plate was placed in an incubator at 37 °C for 2 weeks. Finally, colonies were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet for 20 min, and colonies > 50 μm were counted under a microscope. This experiment was used to analyze the ability of single cells to transform into colonies in 3D under attachment-free conditions. Two layers of agarose gel were prepared: a lower layer with 0.8% gel and an upper layer with 0.4% gel containing 1000 cells in each well of a 96-well plate. Both layers were prepared in serum-free medium. Cells were pretreated with the inhibitor for 30 min, and testing reagents were added on days 1 and 3, and serum-free medium was replenished every 2 days to maintain gel moisture. After 10 days, the spheroid colonies were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet for 20 min. Colonies > 50 μm were counted under a microscope. For the cell invasion assay, a 24-well plate and 8 μm pore-size insert was used. The inserts were pre-coated with 100 µL of diluted matrix Matrigel (#356234, Corning Inc.) overnight. Serum-free medium (2 × 10 4 cells/300 µL) was added to the upper chamber, and 500 µL of serum-free medium was added to the lower chamber. For the inhibition test, cells were pretreated with the inhibitor for 30 min. After 48 h of incubation at 37 °C, cells on the lower surface of the insert were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet. The cells were counted under a microscope. An XTT assay (20-300‐1000 A, BioLegend) was performed in a 96-well plate, and 3000 cells in serum-free medium were seeded in each well. For the inhibition test, cells were pretreated with inhibitor for 30 min. The tested reagent was added once and cultured in a 37 °C incubator for 3 days. Cell viability was determined using 0.04% sulforhodamine B via colorimetric detection of proteins using an ELISA reader at OD 450 nm. For cell xenograft transplantation, 1 × 10 6 human FE25 cells with 200 µg of FF exosomes, with or without inhibitor, in 200 µL saline were injected into the dorsal subcutaneous area and intraperitoneally in immunocompromised mice [NOD/Shi-scid/IL‐2Rγnull (NSG)]. The in vivo tumorigenic appearance was observed using the IVIS imaging system, the image was quantified using Image Wizard software, and the normalized intensity scales and region of interest (ROI) were measured. All experimental procedures were conducted according to the guidelines of the Animal Care and Use Committee of Tzu Chi University (Approval ID: 107‐49). All experiments were performed in duplicate or triplicate and were subsequently repeated in three independent replicates. Data presented as mean ± SD represent three or more independent experiments from technical replicates. Statistical analyses were performed using GraphPad Prism (ver. 5.0c; GraphPad Software), Excel, or SPSS 19.0. Differences between groups were examined using unpaired Student’s t-tests and one‐way analysis of variance. Statistical significance was set at p  < 0.05.

To clarify the role of FF-associated RTK signals in promoting cell transformation, we examined the effects of inhibiting various RTKs, including IGF1R, MET, EGFR, PDGFR, and VEGFR, on the transformation of FTE cells by FF. As shown in Fig.  2 , consistent with our previous studies [ 11 , 13 , 15 ], FF promoted AIG in both FE25 cells (41 ± 1 vs. 1.5 ± 1.0) (Fig.  2 A, B) and FT282-CCNE1 cells (62.5 ± 2.5 vs. 3.0 ± 0.5) (Fig.  2 C, D). Pretreatment with inhibitors of EGFR, IGF-1R, and MET significantly diminished the promoting effects by 96%, 84%, and 67% in FE25 cells and by 77%, 58%, and 64% in FT282-CCNE1 cells, respectively. In contrast, inhibiting PDGFR and VEGFR did not affect the AIG-promoting effects. These results suggest that the EGFR signal is the most potent RTK signal mediating FF promoted transforming activity. Both FE25 and FT282-CCNE1 cells exhibited similar patterns of FF-induced AIG activity, which were inhibited by RTK blockers. Consequently, we selected FE25 cells for further investigation of the role of EGFR signaling in FTE cell transformation. The IC50 of the EGFR inhibitor (Pelitinib or EKB-569) for FE25 cells was determined to be 4.0 µM (Fig. S1 ), and we used a 2.5 µM dose for further examination of the effects of FF-associated EGFR signaling on the tumorigenic behaviors of FTE. Fig. 2 FF exosomes confer most of the clonogenic and transformative activities of FF, which depend on EGFR signaling. ( A ) Schematic flowchart of FF exosome extraction. A spin filter with a 10 kDa cutoff was first used to exclude small-molecule FF fractions (FF < 10 kDa). The remaining FF was subjected to exosome purification using the ExoQuick precipitation reagent. After centrifugation, exosomes were enriched in the pellet (Exo-1). The exosome-poor supernatant (Exo-P) was further treated with a protein G-conjugated CD9 antibody to pull down the residual exosomes to yield exosome-depleted (Exo-D) and residual exosome (Exo-2) fractions. (B-E) Exosome particles in Exo-1 were visualized by electron microscopy (scale bar 100 nm) ( B ), analyzed with NanoSight nanoparticle tracking ( C ), and western blot analysis of CD9, CD81, and HSP 70 markers ( D ). CD9 was further examined in different fractions of FF exosome purification ( E ). ( F ) Calcein AM (1 µM)-stained FF exosomes (Exo-1) were added to FE25 cells for 1–2 h to observe exosome uptake (green arrow) using a fluorescence microscope, combining the differential interference contrast (DIC) and EGFP channels. 400X magnification. Scale bar: 20 μm. ( G , H ) Representative images and quantification results of clonogenicity ( G ) and AIG ( H ) of FE25 cells treated with FF, Exo-1, exosome-depleted FF (Exo-D), and Exo-1 plus EGFR inhibitor (EGFRi), as well as treatment with different concentrations of Exo-1 (50, 100, 150, and 200 µg/mL protein). Scale bar: 100 μm. Data are presented as the mean ± SD ( n  = 3). * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001, comparison as shown in the Figure FF exosomes confer most of the clonogenic and transformative activities of FF, which depend on EGFR signaling. ( A ) Schematic flowchart of FF exosome extraction. A spin filter with a 10 kDa cutoff was first used to exclude small-molecule FF fractions (FF < 10 kDa). The remaining FF was subjected to exosome purification using the ExoQuick precipitation reagent. After centrifugation, exosomes were enriched in the pellet (Exo-1). The exosome-poor supernatant (Exo-P) was further treated with a protein G-conjugated CD9 antibody to pull down the residual exosomes to yield exosome-depleted (Exo-D) and residual exosome (Exo-2) fractions. (B-E) Exosome particles in Exo-1 were visualized by electron microscopy (scale bar 100 nm) ( B ), analyzed with NanoSight nanoparticle tracking ( C ), and western blot analysis of CD9, CD81, and HSP 70 markers ( D ). CD9 was further examined in different fractions of FF exosome purification ( E ). ( F ) Calcein AM (1 µM)-stained FF exosomes (Exo-1) were added to FE25 cells for 1–2 h to observe exosome uptake (green arrow) using a fluorescence microscope, combining the differential interference contrast (DIC) and EGFP channels. 400X magnification. Scale bar: 20 μm. ( G , H ) Representative images and quantification results of clonogenicity ( G ) and AIG ( H ) of FE25 cells treated with FF, Exo-1, exosome-depleted FF (Exo-D), and Exo-1 plus EGFR inhibitor (EGFRi), as well as treatment with different concentrations of Exo-1 (50, 100, 150, and 200 µg/mL protein). Scale bar: 100 μm. Data are presented as the mean ± SD ( n  = 3). * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001, comparison as shown in the Figure To investigate the transforming activity of FF exosomes, we prepared FF fractions with varying numbers of exosomes. As shown in Fig.  1 A, after removing small molecules (< 10 kDa), FF exosomes were extracted using ExoQuick (Exo-1). The exosome-poor supernatant (Exo-P) was then processed with protein G-conjugated anti-CD9 antibodies to pull down any residual exosomes, resulting in a pellet (Exo-2) and an exosome-depleted supernatant (Exo-D). In purified Exo-1, FF exosome particles approximately 100 nm in diameter were first verified by electron microscopy (Fig.  1 B, Fig. S2 ). Further analysis using NTA revealed a mean size of 121.4 ± 4.1 nm, with the exosome concentration in the original volume of FF calculated to be about 2.17 × 10 11 ± 1.3 × 10 10 particles/mL (Fig.  1 C). The exosome markers CD9, CD81, and HSP70 were enriched in the Exo-1 fraction (Fig.  1 D). Based on CD9 abundance, a small number of exosomes (~ 10% of Exo-1) were detected in Exo-2, whereas none were present in Exo-D (Fig.  1 E). We investigated whether the previously identified oncogenic activity of FF is delivered by exosomes and whether EGFR signaling is required for this activity. First, Exo-1 was stained with calcein AM. After incubation for 1 h, FE25 cells exhibited detectable green fluorescence, which further increased after two hours, indicating the successful uptake of exosomes by FE25 cells (Fig.  1 F). The 2D clonogenic analysis revealed that FF exosomes conferred approximately 70% of the clonogenic activity of FF (33.5 ± 1.5 vs. 49 ± 1 colonies), whereas exosome-depleted FF resulted in only 18% of the FF activity (9 ± 1 colonies). The increments were largely diminished by EGFR inhibition, leaving only 4.5 ± 1.5 colonies or 13% of the untreated group (Fig.  1 G). In the AIG assay, the average colony numbers were 49 ± 1, 38 ± 1, and 6.5 ± 0.5 after FF, Exo-1, and Exo-D exposure, respectively. Exo-1 and Exo-D conferred 78% and 13% of the AIG activity of FF, respectively. When the cells were pretreated with EGFR inhibitor, 95% of the AIG activity from Exo-1 was inhibited (49 ± 1 vs. 2.5 ± 0.5). Furthermore, AIG colony numbers were positively correlated with the number of FF exosomes treated (Fig.  1 H, Fig. S3 A). EGFR inhibition almost abolished all AIG activity induced by Exo-1 (Fig.  1 H) or FF (Fig.  2 B). Consistent with our previous findings on FF [ 12 ], FF exosomes promoted AIG in partially transformed late-passage FE25 cells but not in the nontransformed early-passage FE25 cells [ 36 ] (Fig. S4 ). These results indicate that the transformative effects of FF on FTE cells are primarily associated with exosomes and highly dependent on EGFR signaling. As shown in Fig.  3 A-B, the FF exosomes significantly outperformed the FF in promoting FE25 cell proliferation. At 72 h post-treatment, cells treated with FF exhibited a 60% increase in cell number, whereas those treated with FF exosomes displayed a remarkable 150% increase ( p  < 0.001). Notably, the inhibition of EGFR completely negated this increase. Furthermore, the proliferative effect of FF exosomes was found to be dose-dependent and correlated with the concentration of exosomes added (Fig.  3 C and Fig. S3 B). These findings indicate that FF possesses mitogenic cytostatic activities, with the mitogenic effect being particularly enriched in exosomes. Fig. 3 FF exosomes confer the proliferative, migratory, and invasive activities of FF, depending on EGFR signaling. ( A - C ) Images and XTT assay of FE25 cells treated with FF, Exo-1, Exo-D, or Exo-1 plus EGFRi, as well as treated with different concentrations of FF exosomes (50, 100, 150, and 200 µg/mL protein). Scale bar: 50 μm. Data are presented as the mean ± SD ( n  = 4). * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001. ( D - F ) Images and quantification data showing cell migration in the wound healing assay of FE25 cells treated with FF, Exo-1, Exo-D, and Exo-1 plus EGFRi, as well as treatment with different concentrations of FF exosomes (50, 100, 150, and 200 µg/mL protein). Scale bar: 100 μm. Data are presented as the mean ± SD of three microscopic fields for three independent experiments. ( G - I ) Image and quantification data of transwell invasion of FE25 cells treated with FF, Exo-1, Exo-D, and Exo-1 plus EGFRi, as well as treatment with different concentrations of FF exosomes (50, 100, 150, and 200 µg/mL protein). Scale bar: 25 μm. Data are presented as the mean ± SD of three microscopic fields for two independent experiments. * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001 FF exosomes confer the proliferative, migratory, and invasive activities of FF, depending on EGFR signaling. ( A - C ) Images and XTT assay of FE25 cells treated with FF, Exo-1, Exo-D, or Exo-1 plus EGFRi, as well as treated with different concentrations of FF exosomes (50, 100, 150, and 200 µg/mL protein). Scale bar: 50 μm. Data are presented as the mean ± SD ( n  = 4). * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001. ( D - F ) Images and quantification data showing cell migration in the wound healing assay of FE25 cells treated with FF, Exo-1, Exo-D, and Exo-1 plus EGFRi, as well as treatment with different concentrations of FF exosomes (50, 100, 150, and 200 µg/mL protein). Scale bar: 100 μm. Data are presented as the mean ± SD of three microscopic fields for three independent experiments. ( G - I ) Image and quantification data of transwell invasion of FE25 cells treated with FF, Exo-1, Exo-D, and Exo-1 plus EGFRi, as well as treatment with different concentrations of FF exosomes (50, 100, 150, and 200 µg/mL protein). Scale bar: 25 μm. Data are presented as the mean ± SD of three microscopic fields for two independent experiments. * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001 We further investigated the effects of FF exosomes on the migration and invasion of FTE cells. As shown in Fig.  3 D-E and G-H, Exo-1 enhanced the migration and invasion of FE25 cells to levels comparable to those induced by FF. Specifically, FF exosome treatment resulted in 5-fold and 47-fold increases in migration and invasion, respectively. Notably, although EGFR inhibition stopped the migration effect, it only significantly reduced the invasion effect. Figure  3 F, I, and S3 C-D illustrate that the effects of FF exosomes on migration and invasion in FE25 cells were dose-dependent. The tumorigenic effects of FF exosomes and the role of EGFR signaling were assessed using intraperitoneal (IP) and subcutaneous (SC) xenograft co-injection models in NSG mice [ 14 ]. Purified FF exosomes (150 µg Exo-1) were co-injected with 1 × 10 6 luciferase gene-transduced FE25 cells with or without the EGFR inhibitor EKB-569 (2.5 µM). In vivo cell viability and growth were monitored using IVIS. As shown in Fig.  4 , FE25-luc cells co-injected with FF exosomes exhibited a ninefold increase in live cell intensity following SC injection and a sevenfold increase after IP injection. The adding the EGFR inhibitor reduced live cell intensity by 35% (SC) and 65% (IP). These results indicate that FF exosomes may partially through the EGFR signaling pathway to facilitate early peritoneal growth of FE25 cells, raising the potential risk of tumorigenesis. Fig. 4 FF exosomes promote intraperitoneal implantation of FTE cells through EGFR signaling. A total of 1 × 10 6 FE25 cells carrying the luciferase gene (FE25-luc) were IP or SC injected together with 150 µg FF exosomes, with or without EGFR inhibitor (EKB-569, 2.5 µM), into NSG mice. In vivo growth was monitored by IVIS after 2 weeks ( A ); the ROI luciferase intensities in different groups are shown ( B ). ** p  < 0.01, +Exo-1 vs. Vehicle # p  < 0.05, +Exo-1 vs. + Exo-1 + EGFRi FF exosomes promote intraperitoneal implantation of FTE cells through EGFR signaling. A total of 1 × 10 6 FE25 cells carrying the luciferase gene (FE25-luc) were IP or SC injected together with 150 µg FF exosomes, with or without EGFR inhibitor (EKB-569, 2.5 µM), into NSG mice. In vivo growth was monitored by IVIS after 2 weeks ( A ); the ROI luciferase intensities in different groups are shown ( B ). ** p  < 0.01, +Exo-1 vs. Vehicle # p  < 0.05, +Exo-1 vs. + Exo-1 + EGFRi Given the totipotency of the oncogenic activities of EGFR signaling from FF, the exact EGFR ligands involved in FF are of interest. The canonical EGFR ligand EGF was not included in the published proteome of human FF. Instead, several EGF family peptide growth factors, including amphiregulin, epiregulin, TGFα, betacellulin, and versican, are present [ 37 ]. Western blot analysis confirmed that EGF was not present in FF or Exo (Fig. S5 ), whereas amphiregulin, betacellulin, epiregulin, TGFα, and versican were present in FF. Most of the content was present in the Exo-1 fraction rather than in the Exo-P (Fig.  5 A). Fig. 5 FF harbors multiple EGFR ligands, including EGF family growth factors, IGF2, and HGF, all within exosomes. ( A ) Western blot analysis of EGF family growth factors including TGF-α, betacellulin (BCT), epiregulin (EREG), amphiregulin (AREG), and versican (VACN), in FF and Exo-1 and Exo-P fractions. ( B ) An abundance of EGF family growth factors and exosomal HSP70 in Exo-1 before and after treatment with proteinase K (Prot K), Triton-X (Trit X), or both reagents (P + T). ( C ) IGF2 and HGF protein content in Exo-1, FF, and Exo-P FF harbors multiple EGFR ligands, including EGF family growth factors, IGF2, and HGF, all within exosomes. ( A ) Western blot analysis of EGF family growth factors including TGF-α, betacellulin (BCT), epiregulin (EREG), amphiregulin (AREG), and versican (VACN), in FF and Exo-1 and Exo-P fractions. ( B ) An abundance of EGF family growth factors and exosomal HSP70 in Exo-1 before and after treatment with proteinase K (Prot K), Triton-X (Trit X), or both reagents (P + T). ( C ) IGF2 and HGF protein content in Exo-1, FF, and Exo-P Another point of concern is the location of these proteins. They can reside within the exosome as cargo or outside the theme as an associate aggregate [ 38 ]. To differentiate between these two locations, we used Triton-X (T) to perforate the membrane lipid bilayer and proteinase K (P) to digest the exposed proteins. If the protein is inside the exosome, it will be protected from digestion and can only be digested by T + P treatment; if the protein is outside the membrane, it will be digested by P regardless of T. As shown in Fig.  5 B, other than amphiregulin and TGFα which were majorly digested by T + P, the other three EGF ligands and exosomal packaged protein HSP70 [ 39 ] were exclusively digested by T + P, but not P alone. The results indicated that FF-EGFR ligands mostly resided inside exosomes (Fig.  5 B). Previously, two RTK ligands, IGF2 and HGF, were identified in FF and found to have potent oncogenic effects on FTE cells [ 12 – 14 ]. Considering that most of the transformation activities of FF occur in the exosome fraction, we determined whether these oncogenic ligands are also located in the exosomes. As shown in Fig.  5 C, IGF2 and HGF were abundantly present in the purified exosomes of FF, but not in Exo-P, with concentrations higher than in FF. Previous studies of FE25 and FT-CCNE1 cells have demonstrated that FF treatment significantly increased the phosphorylation of AKT, mTOR, NANOG [ 12 ], MAPK, and FAK, whereas it only modestly increased that of STAT3 [ 14 ]. In this study, we examined the five phosphoproteins downstream of EGFR [ 40 ] in FE25 cells after exposure to different FF components. As shown in Fig.  6 A, FF and Exo-1 treatment increased the phosphorylation of EGFR and the downstream signal proteins, including AKT, MAPK, FAK, NANOG, and mTOR. Exo-1 was primarily responsible for the increase, consistent with earlier phenotypic observations. Upon adding the EGFR inhibitor, all of the phosphorylation increases were diminished (Fig.  6 A). These findings suggest that EGFR-mediated transformation signals from FF exosomes operate through the AKT, MAPK, and FAK pathways. In addition to the known IGF axis proteins [ 12 ], EGFR ligands are another source of stemness activation activity in FF. Fig. 6 Signals downstream to FF exosome exposure with or without EGFR inhibition, and summary of RTK signals conferred by FF exosomes. ( A ) Western blot analysis of EGFR phosphorylation and downstream signaling proteins (FAK, MAPK, AKT, mTOR, and NANOG) in FE25 cells with or without FF exosome exposure and EGFR inhibitor pretreatment. Protein intensities relative to the vehicle control are shown. ( B ) Summary of signal transduction and cell transformation effects of the three types of RTK ligands enriched in FF exosomes. EGF-like protein ligands, IGF2, and HGF proteins in FF exosomes induced the transformation phenotypes of exposed FTE cells through different and common pathways. EGF ligands/EGFR and HGF/MET were mediated by FAK, MAPK, and AKT/mTOR to exert migration, invasion, and anchorage-independent growth (AIG) activities, respectively (red and blue arrows). The AIG and stemness activation/clonogenic activities of IGF2/IGF1R are mediated by AKT/mTOR and AKT/NANOG, respectively (yellow arrow) Signals downstream to FF exosome exposure with or without EGFR inhibition, and summary of RTK signals conferred by FF exosomes. ( A ) Western blot analysis of EGFR phosphorylation and downstream signaling proteins (FAK, MAPK, AKT, mTOR, and NANOG) in FE25 cells with or without FF exosome exposure and EGFR inhibitor pretreatment. Protein intensities relative to the vehicle control are shown. ( B ) Summary of signal transduction and cell transformation effects of the three types of RTK ligands enriched in FF exosomes. EGF-like protein ligands, IGF2, and HGF proteins in FF exosomes induced the transformation phenotypes of exposed FTE cells through different and common pathways. EGF ligands/EGFR and HGF/MET were mediated by FAK, MAPK, and AKT/mTOR to exert migration, invasion, and anchorage-independent growth (AIG) activities, respectively (red and blue arrows). The AIG and stemness activation/clonogenic activities of IGF2/IGF1R are mediated by AKT/mTOR and AKT/NANOG, respectively (yellow arrow)

Ovarian high-grade serous carcinoma (HGSC) is often referred to as a silent killer because of the difficulty in detecting its growth in the early stages. Recent research has clarified the etiology of HGSC. Reports indicate that secretory cells in the distal fimbrial region of the fallopian tube are the primary sources of most HGSCs [ 1 , 2 ], and a high lifetime number of ovulations or incessant ovulation is a major risk factor [ 3 – 8 ]. The mechanisms underlying ovulation and the risk of HGSC were clarified when different forms of transforming activities were identified in the ovulatory follicular fluid (FF). Exposing the fallopian tube fimbria to FF during oocyte retrieval induces significant inflammatory and oncogenic changes, including inflammatory responses, reactive oxygen species (ROS) stress associated with DNA double-strand breaks, and notable malignant transformation [ 9 – 14 ]. The transformation activity of FF was clearly demonstrated in in vivo tumorigenesis-inducing experiments. After repeatedly injecting FF into the mammary fat pads of female Trp53-null mice, 50% mice grew local tumors by 6 weeks [ 10 , 15 ]. Testing cells representing different stages of HGSC development revealed that exposure to FF promotes the full course of HGSC development, enhancing all transforming phenotypes, including cell migration, anchorage-independent growth (AIG), cell invasion, peritoneum attachment, anoikis resistance, and cell proliferation [ 9 , 11 , 14 , 16 ]. Two receptor tyrosine kinase (RTK) signals, IGF2/IGF-1R and HGF/MET, were found to be essential for FF-transforming activity; depletion or functional inhibition of any of them abolished tumorigenesis [ 10 , 14 , 15 , 17 , 18 ]. IGF axis proteins, including IGF2, IGFBPs, and PAPP-A, and HGF axis proteins, including coagulation cascade activation proteases, HGFA, and pro-HGF, are abundant in FF. They are activated upon ovulation and exposure to the ovulation wound or fallopian tube fimbria epithelium. The IGF-1R RTK on the surface of FTE transduces the FF-IGF2 signal to promote stemness and clonal expansion through the AKT/NANOG and AKT/mTOR pathways, respectively [ 12 ], whereas the HGF axis proteins are activated by tissue factor-induced coagulation cascade proteases and confer migration, and invasion effects on FTE cells [ 13 ]. Both RTK signals are necessary for the full transforming activity of FF, i.e., AIG and xenograft tumorigenesis [ 14 ]. Interestingly, the transforming activity of FF was durable after ovulation, which was sustained for more than 30 days after drainage into the peritoneal fluid [ 13 ] indicating that RTK ligands are located in extracellular vesicles. Extracellular vesicles (EVs) are lipid bilayer-enclosed particles that carry bioactive molecules. Classified by their size and biogenesis, EV comprise exosomes, microvesicles, and apoptotic bodies. Exosomes are 30–150 nm-sized EVs secreted via the exocytosis of intraluminal vesicles and endosomes. Microvesicles are 40–1000 nm in size and originate from blebbing of the plasma membrane, whereas apoptotic bodies are 50–5000 nm in size and comprise cell debris from apoptosis [ 19 , 20 ]. EVs carry membrane-bound and transmembrane proteins, such as CD63, CD9, CD81, and integrins, which bind to matrix proteins [ 19 ]. Exosomes purified from the FF of different mammals have been shown to promote follicle growth and oocyte maturation [ 21 – 23 ]. Mechanistically, porcine FF exosomes increased the viability, proliferation, and steroidogenesis of granulosa cells through the MAPK/ERK and WNT/β-catenin signaling pathways [ 24 ]. Human FF also contains a large diversity of EVs [ 25 ]. Although FF-EVs seem to play important roles in oocyte development by mediating cell communication [ 26 , 27 ], whether they also mediate the oncogenic activity of FF and play roles in ovulation-induced FTE transformation is unknown. In addition to RTK for IGF2 and HGF, 30–98% of epithelial ovarian cancers (EOCs) have epidermal growth factor receptor (EGFR) overexpression [ 28 – 30 ], which is associated with poor clinical outcomes [ 31 ]. In preclinical studies, small-molecule EGFR inhibitors inhibited the growth of EOC cells and increased their sensitivity to carboplatin treatment. More than 10 clinical trials are currently testing the safety and efficacy of EGFR small-molecule inhibitors as monotherapy or combination therapy for EOC [ 32 ]. In this study, we examined the effects of FF exosomes on AIG, clonogenicity, migration, invasion, proliferation, and in vivo cell growth of immortalized human FTE cells with different extents of transformation. Our data illustrated that in addition to IGF2 and HGF, abundant EGFR ligands are carried by FF exosomes, which are essential for the malignant transformation of FTE cells. This study highlights the role of FF exosomes as carriers of ligands for RTK signaling associated with ovarian HGSC tumorigenesis. The findings are also poised to spark significant interest in future studies on FF exosome-associated carcinogenesis.

This study demonstrated that ovulatory FF exosomes serve as a reservoir of growth factors that deliver EGFR and other RTK signals to transform FTE cells. Although further research is needed to fully elucidate the potential functions and intricate molecular signaling networks involved, these findings highlight the critical influence of ovulation in delivering exosome-protected transforming agents to exposed FTE. The impact of this novel vehicle for delivering oncogenic activity in HGSC development and other cancer types requires further investigation and is highly anticipated.

Extensive studies have indicated a direct link between the lifetime number of ovulatory cycles and EOC risk [ 5 , 41 ], prompting our exploration of carcinogenic factors in ovulation-related FF [ 10 , 12 , 13 , 42 ]. In our previous research, we identified the significant roles of the HGF and IGF2 axis proteins within FF in promoting transformation in FTE cells. HGF activates MET, whereas IGF2 activates IGF1R, both of which belong to the RTK family. In this study, we comprehensively assessed the effects of class-specific RTK inhibitors on the transformative potential of FTE cells induced by FF. Our findings revealed that inhibitors targeting MET and IGF1R reduced FTE cell transformation. Notably, EGFR inhibitors exhibited the strongest effect on FF-induced transformation, while inhibitors of VEGF or PDGF receptors did not impact this transformative potential. The multifaceted transformative activities of follicular fluid (FF) have been previously documented [24355484, 37373301, 26363031, 34688971, 37265438, 38501015]. This study is the first to demonstrate that a significant portion of FF’s transformative effects is linked to exosomes, which promote various processes such as proliferation, clonogenicity, migration, invasion, and the ability of fallopian tube epithelial (FTE) cells to undergo anchorage-independent growth (AIG) and xenograft seeding. FF-derived exosomes are essential for the physiological functions of granulosa and theca cells, as well as for oocyte maturation [ 24 , 43 ][38173013, 38792563, 31052401]. It is reasonable to speculate that FF exosomes carry bioactive substances that not only support normal physiological processes but may also inadvertently facilitate cellular transformation upon interaction with the fimbrial epithelium. The findings of this study highlight the critical role of FF-derived exosomes in the oncogenesis of fimbrial epithelial cells. In addition to known IGF2 and HGF and their corresponding proteins for activation, this study identified five EGF family growth factors (amphiregulin, epiregulin, TGFα, betacellulin, and versican) in FF. All RTK ligands predominantly reside in exosomes. EGF family growth factors are essential for follicle biology [ 44 – 47 ]. Amphiregulin, epiregulin, versican, and betacellulin are potent stimulators of oocyte maturation and cumulus expansion [ 48 ]. Versican processing by ADAMTS1 is important for follicle remodeling before ovulation [ 47 , 49 ]. Moreover, these EGF networks are vital for steroidogenesis, oocyte maturation, and cumulus expansion during folliculogenesis; perturbation of this network in vivo markedly impairs ovulation [ 47 , 48 ]. Our data revealed that FF exosomes exhibited up to 1.5-fold higher activity in promoting cell proliferation than FF alone (Fig.  3 A). We speculate that FF may contain certain detrimental factors that affect cell growth, and this inhibitory effect was nullified in the isolated exosomes. Inflammatory cytokines are associated with the maturation, rupture, and ovulation of follicles [ 50 ]. High free radical levels are present in FF [ 10 , 51 ], and ROS are essential for ovulation [ 51 ]. These factors, including TNF-α, IFN-γ, and free radicals, possess cytotoxic properties. It can be inferred that these are key substances contributing to the cytostatic activities of follicular fluid (FF). In contrast, FF exosomes have a higher proportion of growth factors than that compared to FF (Fig.  5 C), which have led to more pronounced cell proliferation effects. Moreover, although EKB-569 inhibited most of the effects of EGF ligands, some transforming effects of FF exosomes remained detectable. Specifically, approximately 25% of FF exosome-induced FE25 cell invasion activity remained evident after EKB-569 treatment, suggesting that other oncogenic factors, such as HGF, are also important in malignant transformation. Our findings demonstrated that the downstream pathways of EGF-like ligand/EGFR primarily involved the pAKT, pMAPK, and pFAK pathways. In addition to our previous summarization of RTK ligands, IGF2 and HGF, in FF-induced cellular transformation [ 14 ], we here updated the summary by including the EGFR signaling pathways. The EGF ligands/EGFR and HGF/MET are mediated by FAK, MAPK, and AKT/mTOR to exert migration, invasion, and AIG, respectively. Furthermore, the AIG, stemness activation, and clonogenic activities of IGF2/IGF1R and EGF ligands/EGFR were mediated by AKT/mTOR and AKT/NANOG, respectively (Fig.  6 B).

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1