Human hepatic stellate cells (HHSteCs) were obtained from SteCM; ScienCell Research Laboratories and maintained in stellate cell medium (ScienCell Research Laboratories) supplemented with 2% FBS, 1% penicillin/streptomycin solution (ScienCell Research Laboratories) and 1% stellate cell growth supplement (ScienCell Research Laboratories). The human HCC cell lines Hep3B and Huh7 (American Type Culture Collection) were maintained in DMEM (Wako Pure Chemical Industries Ltd.) supplemented with 10% FBS and 1% PenStrep (Thermo Fisher Scientific, Inc.). The human monocytic leukemia cell line THP-1 (American Type Culture Collection) was cultured in RPMI-1640 medium (Wako Pure Chemical Industries Ltd.) supplemented with 10% FBS and 1% PenStrep (Thermo Fisher Scientific, Inc.). All cells were maintained in a humidified incubator with 5% CO₂ at 37˚C. THP-1 cells were induced to differentiate by treating them with 10 mg ml^-l phorbol-12-myristate-13-acetate (Sigma-Aldrich; Merck KGaA) for 3 days. Etoposide (ETP) was purchased from Santa Cruz Biotechnology, Inc. Erlotinib hydrochloride was purchased from Sigma-Aldrich (Merck KGaA).

Cellular senescence was induced by ETP treatment and confirmed by observing p21 and 53BP1 expression in HHSteCs using immunofluorescence assays. A total of 5x10⁴ HHSteCs were mounted on four-chamber slides (Lab-Tek II; Thermo Fisher Scientific, Inc.) and treated with various concentrations of ETP for 3 days. Subsequently, cells were fixed with 4% paraformaldehyde for 30 min at room temperature, permeabilized with ice-cold 70% ethanol and blocked in 1% BSA for 1 h at room temperature. Primary antisera, 1:200 rabbit anti-p21 (cat. no. 29475; Cell Signaling Technology, Inc.) or 1:200 rabbit anti-53BP1 (cat. no. IHC-00001; Bethyl Laboratories, Inc.) were added and the cells were incubated for 1 h at 20-25˚C. After washing the cells with PBS, secondary antisera (AlexaFluor 488-conjugated donkey anti-rabbit IgG; 1:1,000; cat. no. A11008; Molecular Probes; Thermo Fisher Scientific, Inc.) was added to the cells and incubated for 1 h at room temperature. The slides were washed, and coverslips were mounted with DAPI Fluoromount-G (SouthernBiotech). The uptake of EdU was observed in the HHSteCs treated with ETP for 3 days, and for cells left to recover, for another 3 days in normal medium following treatment. EdU staining of the HHSteCs was performed using a Click-iT EdU AlexaFluor 594 imaging kit (cat. no. C10339; Thermo Fisher Scientific, Inc.) for 4 h according to the manufacturer's protocol. Images were acquired using a Keyence All-in-One fluorescence microscope (Keyence Corporation) at x100 magnification. SA-β-gal staining was performed using a Senescence β-Galactosidase Staining kit (Cell Signaling Technology, Inc.) according to the manufacturer's protocol. All assays were performed at least in duplicate.

To collect EVs, 2.5x10⁵ HHSteCs either untreated or pretreated with ETP were seeded in a 100-mm dish and grown in medium containing exo-free FBS (System Biosciences) for 7-10 days. The medium was collected and centrifuged at 300 x g for 10 min and at 16,500 x g for 20 min at 4˚C to remove cells and debris, respectively. After filtration with a 220-nm filter, the supernatant was ultra-centrifuged at 150,000 x g for 120 min at 4˚C. The EV pellet was washed and resuspended in PBS and ultra-centrifuged at 150,000 x g for 120 min at 4˚C. EVs derived from normal cultured HHSteCs and from senescent HHSteCs were termed ‘normal EVs’ and ‘senescent EVs’, respectively. To measure the particle size and number of normal and senescent EVs, 7.5x10⁵ HHSteCs were seeded in a 60-mm dish and grown for 3 days. The EVs were extracted in the same manner as described above and quantified using nanoparticle tracking analysis (NanoSight; Malvern Panalytical).

To examine the incorporation of EVs into hepatoma cells and macrophages, Hep3B and THP-1 cells were treated with EVs labeled with PKH67. EVs were labeled using a PKH67 Green Fluorescent Cell Linker Mini kit (cat. no. MIN167-1KT; Sigma-Aldrich; Merck KGaA) for general cell membrane labeling according to the manufacturer's protocol. Labeled EVs were pelleted by ultracentrifugation twice at 150,000 x g for 120 min at 4˚C to remove excess dye. Subsequently, 5x10⁴ Hep3B and differentiated THP-1 cells each were seeded on four-chamber slides and treated with 2x10⁷ labeled EVs daily for 3 days. PKH67 expression was observed under a Keyence All-in-One fluorescence microscope at x100 magnification. In addition to the unlabeled EVs, PBS without EVs based on the same procedure previously described (to confirm the absence of residual PKH67) was prepared as a negative control. These experiments were performed in at least duplicate.

Comprehensive quantification of growth factors secreted by the cells treated with EVs was performed using multiplex immunoassays. A total of 5x10⁵ Hep3B, Huh7 and differentiated THP-1 cells each were seeded in a 60-mm culture dish and treated with 2x10⁸ EV particles daily for 3 days and the supernatant was subsequently collected. Growth factors were measured using multiplex immunoassays with a Growth Factor 11-Plex Human ProcartaPlex panel (Invitrogen; Thermo Fisher Scientific, Inc.). Epidermal growth factor (EGF) was quantified in the supernatant with a human EGF ELISA kit (cat. no. DEG00; R&D Systems, Inc.) according to manufacturer's protocol.

EGF mRNA expression in differentiated THP-1 cells was analyzed using RT-qPCR. The total RNA was extracted from cells using SuperScript III First-strand Synthesis system (cat. no. 18080051 Thermo Fisher Scientific, Inc.), and cDNA synthesis was performed for 15 min at 42˚C using 1 µg total RNA as a template, RT primer and QuantiTect reverse transcriptase (Superscript III; Thermo Fisher Scientific, Inc.). EGF primers were purchased from Takara Bio, Inc. (cat. no. HA159157) but the sequences of the primers were not disclosed. For qPCR, per a reaction, 1 U LightCycler SYBR Green I Master mix (Roche Diagnostics) was used and the cycling conditions were as follows: Pre-incubation for 5 min at 95˚C; followed by 30 cycles of 10 sec at 95˚C, 10 sec at 60˚C and 10 sec at 72˚C. GAPDH was used as the endogenous control. The sequences of the GAPDH primers were: Forward, 5'-AGCCACATCGCTCAGACAC-3' and reverse, 5'-GCCCAATACGACCAAATCC-3'. Gene expression was calculated using the 2^-ΔΔCq method (30). PCR was performed in triplicate.

To evaluate the impact of EVs on proliferation of hepatoma cells, the viability of Hep3B cells treated with EVs was determined using an MTS assay. A total of 2.5x10³ Hep3B cells were seeded in a 96-well plate and treated with either 1x10⁶ or 3x10⁶ EVs daily for 3 days either with or without differentiated THP-1 cells. Subsequently, 20 µl CellTiter96^® AQueous One Solution Reagent (Promega Corporation) was added to each well. Following incubation for 2 h at 37˚C, the reaction was measured using an automated plate reader (Bio-Rad Laboratories, Inc.) at 490 nm. To evaluate the concentration of erlotinib that could be used whilst maintaining the viability of hepatoma cells, MTS assays were used. A total of 2.5x10³ Hep3B cells were seeded in a 96-well plate and treated with various concentrations of erlotinib for 3 days. MTS assays were performed in at least duplicate.

Data are presented as the mean ± standard deviation of the mean. Multiple comparisons were performed using an ANOVA with a post-hoc Tukey's test. P<0.05 was considered to indicate a statistically significant difference. For multiplex immunoassays which were used as a comprehensive quantification of growth factors, >2-fold difference in secretion was used as the threshold of significance.

HHSteCs were treated with 2.5, 5.0, 25, or 50 µM ETP for 3 days. Alterations in morphology were observed in the treated HHSteCs compared with the untreated cells. HHSteCs treated with 50 µM ETP died within 3 days and the number of cells was notably decreased (Fig. S1). For the HHSteCs treated with 2.5, 5.0 and 25 µM ETP, the expression of p21 (a cell cycle arrest marker), 53BP1 (a DNA damage marker) foci and the uptake of EdU were examined. p21 expression and 53BP1 foci were significantly increased at all three ETP concentrations (Fig. 1A and B), whereas EdU uptake decreased (Fig. 1C) compared with the control. To confirm induction of irreversible cell cycle arrest associated with senescence, HHSteCs treated with each of the ETP concentrations were grown in fresh ETP-free recovery medium for 3 days. EdU uptake was still reduced in HHSteCs treated with 25 µM ETP (Fig. 1D). In contrast, EdU uptake slightly recovered in HHSteCs treated with 2.5 and 5.0 µM ETP. Therefore, 2.5 and 5.0 µM ETP were inadequate for the induction of senescence in HHSteCs. Finally, induction of senescence was confirmed in HHSteCs treated with 25 µM ETP using SA-β-gal staining (Fig. 1E). The proliferation of HHSteCs treated with 25 µM ETP for 3 days was reduced in the recovery medium (Fig. 1F). Thus, 25 µM ETP was used for all subsequent experiments for the induction of senescence in HHSteCs.

After extracting the EVs by ultracentrifugation, the particle size and number of senescent EVs were compared to those of normal EVs using nanoparticle tracking analysis. The size of particles were largely ~120 nm, although some particles were ~200 nm in size. Therefore, it was considered that the extracted EVs were primarily composed of exosomes with a small number of other small vesicles (25,26). The median particle size was 126 nm for normal EVs and 120 nm for senescent EVs (Fig. 2A) and this difference was not significant. To determine the number of EV particles released per HHSteC, the area under the curve (AUC) was calculated using a cell proliferation curve. The cumulative number of EVs in the culture medium was associated with the cumulative number of HHSteCs, although the quantities differed notably between the normal and senescent HHSteC cultures. The AUC of the normal cultured HHSteCs was 1.23x10⁶ cells day^-1, whereas that of the senescent HHSteCs was 7.06x10⁵ cells day^-1 (data not shown). Therefore, it was estimated that there were 2.5x10³ particles produced cell^-1 day^-1 for normal EVs and 4.2x10³ particles cell^-1 day^-1 for senescent EVs. Senescent HHSteCs released ~1.7-fold more EVs per cell than normal cultured HHSteCs, although the significance of this result could not be analyzed statistically.

To confirm whether EVs secreted by HHSteCs were incorporated into both hepatoma cells and macrophages, EVs labeled with PKH67 were added to Hep3B cells and THP-1 cells daily for 3 days. PKH67 expression was observed in both cells (Fig. 2B), which suggests that EVs derived from HHSteCs were incorporated into cells.

To assess the effect of senescent EVs on growth factor secretion from hepatoma cells, Hep3B and Huh7 cells were treated with either normal or senescent EVs daily for 3 days and the secretion of growth factors into the supernatant was measured using a panel of multiplex immunoassays. The difference in secretion of growth factors between normal and senescent EV treatments was <2-fold (Fig. S2). Several studies have shown that TAMs participate in hepatoma cell proliferation via changes in cytokine expression levels (31,32). Thus, the effect of senescent EVs on growth factor secretion from THP-1 cells was assessed. Differentiated THP-1 cells were treated daily for 3 days with either normal or senescent EVs and growth factor secretion was measured. THP-1 cells treated with senescent EVs secreted significantly more EGF compared with those treated with normal EVs and the change was >2-fold (Fig. 3A).

EGF expression in THP-1 cells treated with EVs was further assessed by RT-qPCR and EGF protein-specific ELISA. The levels of EGF mRNA expression in THP-1 cells treated with senescent EVs was significantly higher compared with both the THP-1 cells treated with normal EVs and control (Fig. 3B). To further confirm the results obtained by multiplex immunoassays, EGF secretion was measured using ELISA. EGF secretion from THP-1 cells treated with senescent EVs was also increased compared with THP-1 cells treated with normal EVs and the control (Fig. 3C).

The effect of EVs derived from senescent HHSteCs on hepatoma cell viability was assessed. As shown in Fig. 4A, neither treatment with normal nor senescent EVs affected the proliferation of Hep3B cells. However, both EV treatments significantly increased the viability of Hep3B cells when co-cultured with differentiated THP-1 cells. Notably, this effect was significantly greater with senescent EVs compared with normal EVs (Fig. 4B). To validate the effect of EGF secretion from THP-1 cells on hepatoma cell lines, the cell viability of hepatoma cells in the presence or absence of erlotinib, an EGFR tyrosine kinase inhibitor, was determined. The concentration of erlotinib was set to 2.5 µM to maintain Hep3B cell viability at >80% (Fig. 5A). As shown in Fig. 5B, senescent EVs were more effective at enhancing Hep3B cell viability compared with EVs in the absence of erlotinib; however, this enhancing effect was inhibited in the presence of erlotinib. This suggests that the effect of senescent EVs on the proliferation of hepatoma cells co-cultured with THP-1 cells was dependent on EGF secreted from THP-1 cells.