Extracellular vesicles from senescent hepatic stellate cells promote cell viability of hepatoma cells through increasing EGF secretion from differentiated THP-1 cells

Abstract

Since the discovery of the senescence-associated secretory phenotype, the role of senescent hepatic stellate cells (HSCs) in hepatocellular carcinoma (HCC) development has gained increasing attention. Similar to cytokines, extracellular vesicles (EVs) are essential for intercellular communication. However, the function of EVs derived from senescent HSCs in HCC progression has not been extensively studied. The aims of the present study were to characterize the EVs derived from senescent HSCs and determine their role in the tumor microenvironment. Cellular senescence was induced in human hepatic stellate cells (HHSteCs) with various concentrations of etoposide. Induction was confirmed using EdU staining and 53BP1 and p21 immunostaining. EVs were isolated by ultracentrifugation and analyzed by nanoparticle tracking analysis. Multiplex immunoassays were used to compare the levels of growth factors secreted from hepatoma cell lines and macrophage cells pretreated with EVs derived from senescent HHSteCs (senescent EVs) with those pretreated with EVs derived from normal cultured HHSteCs (normal EVs). Treatment with 25 µM etoposide for 3 days was the most effective at inducing senescence in HHSteCs. This finding was confirmed by induction of irreversible cell-cycle arrest, upregulation of 53BP1 and p21 expression, and increased SA-β-gal staining. Senescent HHSteCs released increased quantities of EV particles compared with normally cultured HHSteCs. Multiplex analysis revealed that there was no difference between hepatoma cell lines treated with normal EVs and those treated with senescent EVs in growth factor secretion. In contrast, the secretion of epidermal growth factor (EGF) was increased by macrophage cells treated with senescent EVs compared with those treated with normal EVs. Furthermore, senescent EVs did not affect the viability of hepatoma cells but increased the viability of hepatoma cells co-cultured with macrophage cells. In conclusion, the release of EVs from senescent HSCs was higher compared with normal HSCs. Furthermore, senescent EVs promoted HCC development by upregulating EGF secretion from macrophages.

Keywords: epidermal growth factor; etoposide; extracellular vesicles; hepatic stellate cells; hepatocellular carcinoma; senescence-associated secretory phenotype.

Figure 1.

Induction of senescence in HHSteCs by ETP treatment. (A) p21 expression (green) was observed in HHSteCs treated with ETP at various concentrations. ETP treatment markedly upregulated p21 expression in HHSteCs compared with the control at all concentrations. (B) 53BP1 foci (green) were observed in HHSteCs treated with ETP at various concentrations. ETP treatment increased 53BP1 foci in HHSteCs compared to the control at all concentrations. (C) EdU uptake (pink) was analyzed in HHSteCs treated with ETP at various concentrations. ETP treatment dramatically reduced EdU uptake by HHSteCs compared with the control at all concentrations. (D) EdU uptake was analyzed in HHSteCs grown in fresh medium for 3 days following treatment with ETP at various concentrations. EdU uptake recovered in the HHSteCs treated with 2.5 and 5.0 µM ETP, but not in HHSteCs treated with 25 µM ETP. (E) SA-β-gal staining was observed in untreated HHSteCs and HHSteCs treated with 5.0 and 25 µM ETP. Only HHSteCs treated with 25 µM ETP exhibited a notable induction of senescence, as shown by the increase in staining. (F) Growth curves in fresh medium were compared between Normal HSC or Senescent HSCs for 3 days. No proliferation of senescent HSCs was observed in fresh medium. ^*P<0.05. ETP, etoposide; HHSteCs, human hepatic stellate cells; Normal HSC, HHSteCs that had been cultured in normal medium; Senescent HSC, HHSteCs that had been cultured in medium containing 25 µM ETP.

Figure 2.

Particle size of EVs secreted from HHSteCs and their incorporation into hepatoma cells. (A) Nanoparticle tracking analysis of normal and senescent EVs. Particle size distributions were nearly identical between normal and senescent EVs, and the majority of particles were ~120 nm in size. (B) Hep3B cells and THP-1 cells were treated with PKH67-labelled EVs daily for 3 days and observed by fluorescence microscopy to examine EV incorporation. Magnification, x100. Unlabeled EVs and PBS without EVs (for confirmation of the absence of residual PKH67) were used as the NCs. PKH67 expression (green) was observed in only the Hep3B cells treated with PKH67-labeled EVs. EV, extracellular vesicle; NC, negative control; HHSteCs, human hepatic stellate cells; normal EVs, EVs derived from normal HHSteCs; senescent EVs, EVs derived from senescent HHSteCs.

Figure 3.

EGF upregulation in THP-1 cells treated with EVs secreted from senescent HHSteCs. (A) Comprehensive quantification of growth factors secreted by differentiated THP-1 cells treated with EVs was performed using multiplex immunoassays. THP-1 cells treated with senescent EVs secreted increased quantities of EGF compared with normal EVs, and the increase was >2-fold. (B) EGF mRNA expression levels were compared between the different treatments. THP-1 cells treated with senescent EVs had higher EGF mRNA expression levels compared with THP-1 cells treated with normal EVs and the control. (C) EGF secretion was examined in control THP-1 cells, THP-1 cells treated with normal EVs and THP-1 cells treated with senescent EVs. THP-1 cells treated with senescent EVs secreted significantly more EGF compared with the control or normal EV treated THP-1 cells. ^*P<0.05. EGF, epidermal growth factor; EV, extracellular vesicle; HHSteCs, human hepatic stellate cells; normal EVs, EVs derived from normal HHSteCs; senescent EVs, EVs derived from senescent HHSteCs.

Figure 4.

Impact of EVs on the proliferation of hepatoma cells alone or those co-cultured with THP-1 cells. To determine the effect of EVs on hepatoma cell proliferation, the viability of Hep3B cells treated with EVs was evaluated by MTS assays. (A) In the absence of THP-1 cells, neither normal EVs nor senescent EVs affected Hep3B cell viability. (B) Upon co-culture with THP-1 cells, both normal EVs and senescent EVs significantly increased Hep3B cell viability. Treatment with a higher quantity of senescent EVs significantly increased Hep3B cell viability compared with treatment with an equal quantity of normal EVs. ^*P<0.05. ^#P<0.05 vs. co-culture with THP-1 cells without EV treatment. N.S., not significant; EV, extracellular vesicle; normal EVs, EVs derived from normal human hepatic stellate cells; senescent EVs, EVs derived from senescent human hepatic stellate cells; +, 1x10⁶ added daily; ++, 3x10⁶ added daily.

Figure 5.

Effect of EVs derived from senescent HHSteCs co-cultured with THP-1 cells on the proliferation of hepatoma cells in the presence of erlotinib, an epidermal growth factor receptor inhibitor. To determine the involvement of EGF secreted from senescent HHSteCs in hepatoma cell proliferation upon co-culture with THP-1 cells, the viability of Hep3B cells co-cultured with THP-1 cells was evaluated in the absence or presence of erlotinib using MTS assays. (A) The effect of erlotinib on the inhibition of cell viability was examined at each concentration. Cell viability was maintained at >80% with 1.0 and 2.5 µM erlotinib treatment. (B) Cells were treated with two different quantities of EV particles daily for 3 days. Senescent EVs were more effective at enhancing Hep3B cell viability compared with normal EVs in the absence of erlotinib; however this effect was abrogated in the presence of erlotinib. ^*P<0.05. N.S., not significant; EV, extracellular vesicle; HHSteCs, human hepatic stellate cells; EGF, epidermal growth factor; +, 1x10⁶ added daily; ++, 3x10⁶ added daily; normal EVs, EVs derived from normal human hepatic stellate cells; senescent EVs, EVs derived from senescent human hepatic stellate cells.