Author Correction: N-terminal gelsolin fragment potentiates TRAIL mediated death in resistant hepatoma cells

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.

Hepatocellular carcinoma (HCC) is a malignancy of worldwide significance and has become increasingly important in the United States. Novel pharmacological modality is urgently needed for HCC treatment. TRAIL may be of potential use as an anticancer drug for tumor selectivity, minimal side effect in animal models, and promising results from phase I/II clinical studies 1 . TRAIL initiated intracellular apoptosis signal transduction involves the TRAIL-death receptors (DR4 and DR5), Fas-associated protein with death domain (FADD) and caspase signaling 2 . TRAIL can activate the extrinsic pathway of cell death by binding to the death receptors, DR4 and DR5. The apoptosis signal of TRAIL may be amplified by mitochondria, which is regulated by members of the Bcl-2 family. However, HCC cells exhibit a major resistance to TRAIL-induced cell death.
Due to varying factors within individual established tumors leading to resistance to TRAIL mediated growth inhibition, the antitumor effect of TRAIL as a single agent is limited. Cytotoxic drugs, such as doxorubicin, methotrexate and others induce apoptosis along with TRAIL 3 . Several mechanisms work for cytotoxic drugs sensitizing tumor cells for TRAIL-induced apoptosis. Among them, p53 is activated in tumor cells by several cytotoxic drugs and mediates gene regulation, apoptosis and cell cycle arrest. Several proteins mediate TRAIL-induced apoptosis, including TRAIL receptor 2 or DR5 as p53 target gene. Therefore p53-mediated gene regulation is a mechanism for mediating apoptosis of cytotoxic drugs and TRAIL 4 . Activation of the PI3K/Akt pathway is associated with tumorigenesis and resistance to apoptosis, and inhibition of Akt activation also enhances TRAIL mediated cell death [5][6][7] .
Our previous study suggested that conditioned medium (CM) from immortalized human hepatocytes (IHH) induced apoptosis in human hepatic stellate cells (LX2). Peptide mass fingerprinting of a purified soluble mediator from CM indicated that gelsolin fragments may play a role in LX2 apoptosis 8 , and similarly modulated MAPK/Akt/Mdm2/Bcl2, and enhanced Bax, in the absence of TRAIL (unpublished observations). Further studies indicated that the N-terminal gelsolin 1-70 fragment also induces LX2 cell death in the absence of TRAIL and decreases Bcl2 expression. Gelsolin, a multifunctional actin-binding protein, is downregulated in several types of tumors and its abnormal expression is one of the most common defects noted in invasive breast carcinoma 9 . Loss of gelsolin, a tumor suppressor, is one of the most frequently occurring molecular defects in breast cancers of diverse etiologies in human, mouse, and rat 10 . CM increased the expression of TRAIL receptors on LX2 surface, and induced apoptosis by a caspase dependent mechanism 11 . Gelsolin is secreted from several mammalian cell types. Originally defined by its interactions with actin, plasma gelsolin circulates in mammalian blood at concentrations of 200-300 µg/ml [12][13][14][15] . An earlier study identified an N-terminal gelsolin fragment obtained by caspase 3 mediated cleavage in response to IFN-γ and TNF-α exposure 16 . This fragment reduced cell viability in a manner similar to our previous work 8,11 . Further analysis determined that this activity was restricted to a region encompassing amino acids 1-70 in the gelsolin sequence 11 , and antibody against a linear B-cell epitope from this region inhibits stellate cell death (unpublished observation). This fragment upregulated TRAIL-R1/TRAIL-R2, and involved caspase 3 activation. The apoptotic activity of the N-terminal gelsolin fragment was restricted to activated, not quiescent, stellate cells indicating its potential application as an anti-fibrotic agent.
Sorafenib, a multikinase inhibitor, improves overall survival in patients with advanced HCC 17 . However, there is urgent need for additional pharmacological modalities for HCC. Gelsolin has a tumor suppressor activity in breast cancers 9,16 , although the role of gelsolin in HCC remains unknown. Here, we examined whether gelsolin can potentiate TRAIL mediated cell death in resistant human hepatoma cells. Our findings indicated that the gelsolin fragment sensitizes transformed hepatocytes to TRAIL-mediated apoptosis through modulation of cell survival pathways. The results suggested that the combination of TRAIL and the N-terminal gelsolin fragment may be effective for treatment of complexities of HCC associated desmoplasia.

Results
Conditioned medium from immortalized hepatocytes potentiates TRAIL mediated cell death in transformed cells. Normal human hepatocytes do not express detectable TRAIL receptor 1 (TRAIL-R1), and only a limited level of receptor 2 (TRAIL-R2) 18 . To evaluate the mechanisms of tumor resistance to TRAIL and the ability of gelsolin to sensitize cells to TRAIL mediated cell death, we investigated human hepatoma (HepG2) and hepatocellular carcinoma (Huh7) cells exhibiting reduced sensitivity to TRAIL 19 . Previous studies have established reduced sensitivity of HepG2 and Huh7 cells to TRAIL [20][21][22][23][24] and subsequent analysis in our laboratory concurred with these results (data not shown) and allowed for selection of the TRAIL dose used in each cell line. Treatment of the cells with CM from IHH or TRAIL (25ng/ml for HepG2 and 50ng/ml for Huh7) alone induced modest (10-20%) cell death (Fig. 1A). However, a combination of both CM and TRAIL reduced cell viability in HepG2 (~90%) and Huh7 (~70%) cells, indicating that CM sensitizes resistant hepatocytes to TRAIL mediated death.
We analyzed the expression of TRAIL-R1 in HepG2 and Huh7 (Fig. 1B,C) cells. A modest increase in TRAIL-R1 was observed in HepG2 cells after treatment with CM/TRAIL. On the other hand, Huh7 cells did not display an increase in TRAIL receptor upon treatment with CM, and had a similar high expression level in the untreated control. These results suggest that the expression of TRAIL receptor differs between these two cell lines, and the induction of cell death by the combination of CM and TRAIL may not be identical. Further, a decrease in TRAIL receptor expression in treated Huh7 cells may indicate that modulation of this receptor does not relate to TRAIL mediated cell death. We further evaluated the effects CM and the N-terminal gelsolin 1-70 fragment upon cell survival pathways independent of TRAIL receptor enhancement as described below.

Combination of CM and TRAIL impairs Akt, Erk, or MAPK activation in transformed cells.
The activation of the phosphoinositide 3-kinase (PI 3-kinase)-Akt signaling cascade protects cells from apoptosis 25 . In addition, the triggering of stress-activated kinases, concomitant with the inhibition of the extracellular signal-regulated kinase (Erk) pathway, are observed in cells undergoing programmed cell death. The PI3K and MAPK pathways are associated with cell survival in response to TRAIL mediated apoptosis. Treatment with CM reduced levels of Akt phosphorylated at Ser473 in both HepG2 and Huh7 cells ( Fig. 2A,B). However, the levels of Akt remained constant upon treatment of HepG2 and Huh7 cells with TRAIL. Further, co-treatment with CM and TRAIL led to greatly inhibit Akt activation. Next, we analyzed expression of Erk in response to TRAIL, CM, and a combination of both. The use of TRAIL individually exhibited a moderate effect on the activation of Erk, while Erk activation was decreased in CM treated cells (Fig. 2C,D). However, the use of both in concert led to a significant decrease in the phosphorylation of Erk protein, with the onset of cell death. Modulation of both Akt and Erk activation may contribute to the loss of TRAIL resistance.

CM in the presence of TRAIL modulates p53/Mdm2 activation in transformed hepatocytes.
Since p53 and Akt are major but opposing players in signaling pathways that determine cell survival, it is conceivable that there may be cross-talk between these two proteins to enable integration of signaling information towards death or survival. HepG2 cells carry wild-type p53, whereas Huh7 cells have point mutations in p53 at codon 220 resulting in substitution from tyrosine to cysteine to inhibit functional activity 26 . We analyzed the potential for CM/TRAIL to affect p53 activation status. Phosphorylation of p53 on the Serine15 residue was apparent when HepG2 cells were treated with both CM from IHH and TRAIL, without an increase in total p53 content (Fig. 3A). In similar fashion, p53 phosphorylation was also noted in the p53 defective Huh7 cell line, indicating that the associated phosphorylation was conserved even in the absence of p53 functionality (data not shown). p53 is the main transcriptional regulator of p21, and p21 has been observed to inhibit TRAIL mediated apoptosis 27 . p21 levels were increased in HepG2 cells upon TRAIL exposure, but this effect was reduced in the presence of IHH CM. Modulation of p53 activation in HepG2 cells may, in part, contribute to gelsolin and TRAIL mediated cell death, and may partly explain decreased cell death apparent in Huh7 cells as compared to HepG2. However, a p53 negative hepatoma cell line, Hep3B, exhibited significant cell death in the presence of conditioned medium alone (data not shown). This may further indicate that there is more than one cellular pathway affected by CM leading to cell death.
SCIentIfIC RePoRTs | 7: 12803 | DOI:10.1038/s41598-017-13131-7 The major physiological regulator of p53 function is the Mdm2 oncoprotein. Mdm2 through its action as a p53-specifc E3 ubiquitin ligase ubiquitinates p53 and targets the protein for rapid proteasomal degradation [28][29][30] . When Akt becomes active, phosphorylation occurs at an Akt consensus site (Serine166) within the Mdm2 protein modulating cellular interactions which promote p53 degradation 31 . Our, results suggested that a combination of CM and TRAIL led to a decrease in activated Akt (shown previously in Fig. 2A,B). Densitometric scanning for comparison of protein expression levels are shown at the bottom of each panel. Therefore, Akt may modulate the p53 pathway by altering the intracellular translocation of Mdm2 31 . Consequently, we examined the phosphorylation of Mdm2 at Ser166. The treatment of both HepG2 and Huh7 cells with CM and TRAIL displayed a much reduced level of activated Mdm2 (Fig. 3B and data not shown). Phosphorylations of Mdm2 at Serine166 and p53 at Serine15 are associated with cellular interactions between these two proteins, leading to modulation of p53 activity. Control cells displayed significant association between p53 and Mdm2 when exposed to TRAIL alone (Fig. 3C). In contrast, the association between p53 and Mdm2 was not apparent in cells treated with both CM and TRAIL. These results indicate that p53 functional activity may contribute to TRAIL mediated cell death in the presence of gelsolin, but is not the only mechanism by which decreased cell viability may occur via these ligands.  Here, we analyzed the ability of a combination of IHH CM and TRAIL to modulate the expression of survivin and Bcl-2, as well as the pro-apoptotic BH3 family member, Bax. The expression of survivin or Bcl-2 was decreased in HepG2 cells treated with both CM and TRAIL, with very little change in the expression level of Bax (Fig. 4A-C). Densitometric scanning for comparison of protein expression levels are shown at the bottom of each panel. Further, treatment of Huh7 cells with either CM or TRAIL exhibited a modest change in each of these proteins (Fig. 4D-F). In contrast, the use of both ligands displayed a significant decrease in both survivin and Bcl-2 expression (Fig. 4D,E), but did not affect the level of Bax expression in Huh7 cells (Fig. 4C).

Treatment of cells with N-terminal gelsolin fragment decreases HCC cell proliferation.
We examined HepG2, Huh7 and Hep3B cell lines following treatment with N-terminal gelsolin peptide  . Each cell line exhibited minimal loss in cell viability, with the most significant level (~30%) exhibited by Hep3B (Fig. 5A). Further analysis was undertaken to determine if a reduction in cell viability mirrored that seen with CM and Figure 2. CM in association with TRAIL inhibits Akt and Erk activation. Akt and Erk activation was analyzed in HepG2 (panels A and C) and Huh7 (panels B and D) cells. CM was added on cells 24 hours prior to the addition of TRAIL. Densitometric scan data is provided. Each panel is representative of three independent experiments. Note that cropped gel images are used and the gels were run under the same experimental conditions. TRAIL treatment in the transformed hepatocyte cell lines in the presence of the gelsolin 1-70 peptide and TRAIL. Cells were treated with a fixed concentration of TRAIL and varied concentrations of the gelsolin 1-70 peptide (0.3-5.0 ug/ml). A significant reduction in cell viability (>70%) was observed in HepG2 and Huh7 cells (Fig. 5B). However, a p53 null cell line, Hep3B, exhibited minimal effect with the addition of TRAIL only at the highest concentration of the gelsolin fragment used in this assay. TRAILR1 modulation for HepG2 and Huh7 was similar to that seen using CM/TRAIL (Fig. 5C). PARP cleavage to a 89 Kd fragment is a marker for apoptotic cell death. Western blot analysis for cell death associated protein PARP/cleaved fragment in HepG2 and densitometric scanning results are shown (Fig. 5D).
We analyzed selected proteins to identify key targets for the gelsolin fragment leading to TRAIL sensitization of HepG2 cells as a representative cell line. Phosphorylated Mdm2, Akt and Erk were significantly reduced in CM or gelsolin 1-70 treated cells (Fig. 5E). Further, survivin levels were also reduced in the presence of CM or gelsolin 1-70 /TRAIL. Densitometric scanning for comparison of protein expression levels are shown at the bottom of each panel. These results indicate further that the gelsolin 1-70 fragment exerts significant influence upon transformed cells, which are resistant to TRAIL potentiating cell death.

Activation of Akt by sc-79 enhanced HepG2 cell viability to treatment with the N terminal gelsolin fragment and TRAIL.
To activate the Akt pathway in the presence of gelsolin 1-70 /TRAIL, we used the . CM and TRAIL reduces association between p53 and Mdm2. p53 and Mdm2 phosphorylation was analyzed in CM, TRAIL, and CM + TRAIL treated HepG2 cells (panels A and B). Cellular phosphorylation status of p53 at Serine15 residue was apparent when cells were exposed to both CM and TRAIL, but total p53 levels were not affected. Phosphorylated Mdm2 was reduced in the presence of both CM and TRAIL (panel B). Densitometric scan data is provided. Note that cropped gel images are used and the gels were run under the same experimental conditions. Subcellular localization of p53 and Mdm2 is shown by immunofluorescence (panel C). Co-localization of p53 and Mdm2 was noted in HepG2 control cells, and their association noted in the presence of TRAIL. This interaction was not observed in cells treated with CM and TRAIL when examining multiple fields. Each panel is representative of three independent experiments.
SCIentIfIC RePoRTs | 7: 12803 | DOI:10.1038/s41598-017-13131-7 Akt activating agent; Sc-79, which specifically binds to the PH domain of Akt, Sc-79-bound Akt adopts a conformation favorable for phosphorylation by upstream protein kinases 33 . Activation of the Akt pathway is associated with a decrease in sensitivity to TRAIL treatment 34 . As above, treatment with gelsolin 1-70 /TRAIL led to a decrease in HepG2 cell viability, as compared to untreated cells and those treated with gelsolin peptide alone. The addition of SC79 in cells treated with gelsolin 1-70 /TRAIL restored cell viability (Fig. 6A). Expression of activated Akt in the presence of sc-79 and gelsolin 1-70 /TRAIL are shown by Western blot analysis and densitometric scanning (Fig. 6B). Interestingly, the expression of activated Erk was also observed after this treatment. This data indicates that the loss of activated Akt may play an important role in the observed decrease in cell viability which TRAIL treated HepG2 cells undergo in the presence of the gelsolin fragment.

Inhibition of p53 expression by siRNA enhanced cell viability in p53 positive HepG2 cells treated with the gelsolin fragment and TRAIL.
As the p53 null cell line; Hep3b, displayed minimal effect by the co-treatment of the gelsolin fragment and TRAIL, we further analyzed the contribution p53 makes in mediating this loss in cell viability. Cells treated with gelsolin 1-70 fragment displayed increased viability when prior treated with a siRNA to p53 (Fig. 7A), even at increased concentration of TRAIL. Analysis of protein expression by Western blot indicated that not only was p53 expression eliminated, but the phosphorylation of p53 at Ser15 was not present (Fig. 7B). Interestingly, the addition of the sc-79 as an Akt activating agent led to an increase in p21 expression in cells treated with p53 siRNA. Similarly, a constitutively active form of Akt has previously been noted to induce p21 expression in human endothelial cells 35 . Thus, the role of p53 in the mediation of cell death in p53 competent, TRAIL resistant cells occurs when co-treated with the gelsolin 1-70 fragment.

Discussion
We have previously identified that an N-terminal gelsolin fragment as a soluble mediator secreted from HCV core immortalized primary human hepatocytes mediates enhanced expression of the death receptors, TRAIL-R1 and TRAIL-R2, and induces apoptosis in LX2 cells 8,11 . Cytokine signaling through TNF receptors play a critical role in host defense by selectively eradicating infected and malignant cells from healthy cell populations 36 , and is hence of interest as a potential therapeutic agent. TRAIL-R1 and TRAIL-R2 facilitate the selective elimination of malignant cells through the induction of apoptosis via a caspase-dependent pathway 5 . A large number of tumorigenic cell lines suppress TRAIL mediated cell death using distinct mechanisms. Here, we identified that the N-terminal gelsolin fragment was able to synergistically aid TRAIL mediated cell death by modulating the MAPK/Akt and Bcl2/survivin signaling molecules. Previously, Huh7 and HepG2 cells exhibited minimal cell death in response to TRAIL or gelsolin 8,18,19 . To clarify the role of the N-terminal gelsolin sequence, a synthetic peptide comprising the amino acid residues 1-70 was utilized and was able to induce decreased cell proliferation in transformed hepatic cell lines treated concurrently with TRAIL.
We observed in this study that the gelsolin fragment or CM reduced the level of activated Akt with increased cell death. The activation of the MAPK/Erk pathway has been associated with the induction of Erk1/2 signal resulting in inhibition of caspase activation in response to TRAIL-R stimulation 37 . However, the addition of gelsolin containing CM suppressed activation of Erk in the presence of TRAIL. These results highlight the ability of gelsolin to stimulate TRAIL mediated induction of apoptosis by reducing the cells cyto-protective kinase pathways. Additionally, use of CM or the gelsolin fragment in the presence of TRAIL reduced the phosphorylation of Akt or ERK.
A number of studies have shown that treatment of tumor cells with chemotherapeutic agents and irradiation results in sensitization of TRAIL-resistant tumor cell lines [38][39][40] . However, the mechanisms leading to TRAIL sensitivity are controversial 38,[41][42][43] . Here, we demonstrated that an N-terminal gelsolin fragment and TRAIL have a synergistic effect of inhibiting cell growth against TRAIL resistant transformed human hepatic cell lines. We observed that cell growth regulation by TRAIL and the gelsolin fragment are associated with multiple cellular mechanisms.
Sorafenib, a multi kinase inhibitor, can sensitize cancer resistant cells to cell death through the regulation of Stat3 and the downstream cytoprotective proteins, Mcl1 and survivin 44,45 . The use of sorafenib improves overall survival in patients with advanced HCC 17 . Activation of the Akt signaling pathway induces resistance to Sorafenib in hepatocellular carcinoma cells 46 . A co-culture of Huh7 with LX-2 cells induced phosphorylation of met, Akt, ERK and Stat3 associated with Sorafenib resistance 47 . The activation of the MAPK/Erk pathway associates with the induction of Erk1/2 resulting in inhibition of caspase activation in response to TRAIL-R stimulation 37 . These results highlight the ability of gelsolin to stimulate TRAIL mediated induction of apoptosis by reducing cyto-protective kinase pathways in cells, with the previously reported potential to eliminate activated stellate cells in culture. We have observed that Akt and Bcl2 levels were significantly reduced with the concomitant treatment of Huh7 and HepG2 cells with the gelsolin fragment and TRAIL. Successful cancer treatment may be associated with understanding the cellular mechanisms that distinguish TRAIL-resistant from TRAIL-sensitive cells, and the ability to include treatment regimens that can overcome TRAIL resistance in vivo. HCC is one of the most common carcinomas, and options for successful therapy are limited due to chemotherapy resistance.
Targeting the stromal environment for cancer treatment is an emerging field. The Stromal environment constitutes the connective tissue framework of the liver and kidney distinguishing tumor from the tissue performing normal cell functions. We anticipate gelsolin will potentiate TRAIL mediated cell death in desmoplasia where resistant stromal cells are normally difficult to treat. Indeed, the LX2 model utilized by us was resistant to TRAIL mediated cell death, but exhibited decreased viability in the presence of the gelsolin fragment. The use of combination therapy focused upon the N-terminal gelsolin fragment and TRAIL could potentially be highly effective as a therapeutic agent together, or included in a treatment course with other therapeutics. Figure 7. Inhibition of p53 expression decreased the loss in cell viability in p53 positive cells treated with an N-terminal gelsolin 1-70 fragment and TRAIL. Treatment of HepG2 cells with siRNA to p53 prior to exposure to the N-terminal gelsolin 1-70 fragment and TRAIL enhanced cell viability after 48 hours of treatment (panel A). Treatment of HepG2 cells with siRNA to p53 prior to exposure to the N-terminal gelsolin 1-70 fragment and TRAIL inhibited phosphorylation of p53 at S15 and enhanced p21 expression by Western blot analysis (panel B). Densitometric scanning results are shown on the right. Each panel is representative of three independent experiments and mean values ± SD are shown as error bars. Note that cropped gel images are used and the gels were run under the same experimental conditions.