Human renal carcinoma cells respond to Newcastle disease virus infection through activation of the p38 MAPK/NF-κB/IκBα pathway

Abstract

Purpose

Newcastle disease virus (NDV) is an oncolytic virus that is known to have a higher preference to cancer cells than to normal cells. It has been proposed that this higher preference may be due to defects in the interferon (IFN) responses of cancer cells. The exact mechanism underlying this process, however, remains to be resolved. In the present study, we examined the antiviral response towards NDV infection of clear cell renal cell carcinoma (ccRCC) cells. ccRCC is associated with mutations of the von Hippel-Lindau tumor suppressor gene VHL, whose protein product is important for eliciting cellular responses to changes in oxygen levels. The most common first line treatment strategy of ccRCC includes IFN. Unfortunately, most ccRCC cases are diagnosed at a late stage and often are resistant to IFN-based therapies. Alternative treatment approaches, including virotherapy using oncolytic viruses, are currently being investigated. The present study was designed to investigate the mechanistic pathways underlying the response of ccRCC cells to oncolytic NDV infection.

Methods and results

We found that NDV induces activation of NF-κB in ccRCC cells by inducing phosphorylation and subsequent degradation of IκBα. IκBα was found to be phosphorylated as early as 1 hour post-infection and to result in rapid NF-κB nuclear translocation and activation. Importantly, p38 MAPK phosphorylation was found to occur upstream of the NDV-induced NF-κB activation. Restoration of VHL in ccRCC cells did not result in a reduction of this phosphorylation. A similar phenomenon was also observed in several other cancer-derived cell lines.

Conclusion

Our data provide evidence for involvement of the p38 MAPK/NF-κB/IκBα pathway in NDV infection and subsequent induction of apoptosis in ccRCC cells.

Introduction

Newcastle disease virus (NDV) is one of the prime candidates for anticancer virotherapy [1]. Cancer cells have amply been shown to have a higher sensitivity to NDV-mediated killing than normal cells, due in part to defects in an antiviral interferon (IFN) response of the cancer cells [1–3]. Most cancer cells produce only IFN-β in response to NDV infection, whereas normal cells produce IFN-α/β after such infection [4, 5]. The antiviral responses observed included induction of type I IFN that is accompanied by activation of a critical transcription factor, i.e., nuclear factor-kappaB (NF-κB). Activation of pathways that are associated with these molecules promotes cells to undergo an antiviral state via the involvement of various response pathways, including the Janus kinase/signal transducer and activator (JAK-STAT) pathway [reviewed in [6]. Recently, we showed that NDV infection in clear cell renal cell carcinoma (ccRCC)-derived cell lines leads to the production of IFN-β via STAT1 activation [5]. IFN-β, but not IFN-α production was found to be associated with increased STAT1 phosphorylation in the infected cells. High levels of IFN-β in the supernatant culture media of infected cells also led to a prolonged STAT1 phosphorylation. Others have shown that regulation of IFN-β production by NDV in mouse embryonic fibroblasts (MEF) involves NF-κB, specifically during early phases of virus infection [7].

These observations led us to consider the possibility of manipulating the exclusive IFN-β induction by NDV [5] as a potential strategy to boost the efficacy and safety of NDV as an oncolytic agent in clinical settings. This possibility may ultimately be explored when a detailed understanding of the early cellular responses to NDV infection becomes available. Given the central role of NF-κB in the type I IFN antiviral innate immune response, we sought to investigate the correlation between IFN-β production, NF-κB pathway activation and NDV-induced cytolysis in human ccRCC, using the ccRCC-derived cell line 786-O as a model system.

Materials and methods

Cell lines, Newcastle disease virus and culture conditions

The 786-O (human renal clear cell carcinoma), MCF7 (human mammary gland adenocarcinoma) and Saos-2 (human osteosarcoma) cell lines were obtained from the American Type Culture Collection (ATCC). The 786-VHL cell line was developed as previously reported [5]. The cells were grown at 37 °C in Dulbecco’s modified Eagle’s medium (DMEM; PAA, Pasching, Austria) supplemented with 10 % fetal bovine serum (FBS; PAA, Pasching, Austria) in a humidified CO₂ incubator. A local velogenic Newcastle disease virus strain, AF2240, was propagated as previously reported [5, 8, 9]. In cases where JSH-23 (Merck KGaA, Darmstadt, Germany) was used, the drug (60 μM) was added 1 h prior to NDV infection. NDV infection of cells was performed at 0.1 or 1.0 multiplicity of infection (MOI; [5, 10]). Zero hour post-infection (hpi) represents the time point immediately after virus addition to cell cultures. Cell viability and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining assays were performed as reported previously [5].

Interferon-β production measurement

Cells (4.73 × 10⁵) were grown overnight in 6-well plates and then infected with 1.0 MOI of NDV. After pre-selected time points post-infection, cell culture supernatants were harvested and centrifuged at 1780×g for 5 min. IFN-β levels in the supernatants were measured using the VeriKine Human Interferon-Beta ELISA kit (PBL interferon source, Piscataway, NJ, USA) as reported previously [5].

NF-κB activity measurement

Cells (4.5 × 10⁴) were seeded into 24-well plates and incubated overnight. Next, they were co-transfected with pGL4.32 [luc2P/NF-κB-RE/Hygro] (Promega, Madison, WI, USA) and pRL-CMV (Promega, Madison, WI, USA) at a ratio of 10:1 using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). After a 24 h incubation period, the cells were infected with NDV. As a positive control for NF-κB activation, bacterial lipopolysaccharide (LPS, 10 μg/ml; Sigma Aldrich, St. Louis, MO, USA) was used. To inhibit LPS-induced NF-κB activation, JSH-23 (60 μM) was added 1 h following LPS treatment initiation. After 1 h, cell culture media were replaced with fresh media and the samples were incubated for another 24 h prior to harvesting. The relative luciferase activity (pGL4.32) in the cells was measured using the Dual Luciferase Reporter Assay System (Promega, Madison, WI, USA), in accordance with the manufacturer’s instructions.

Protein harvesting and Western blotting

Cells (3.75 × 10⁶) were grown overnight in T-75 flasks, followed by NDV infection and harvesting at different time points post-infection. For total cell lysate preparations, the harvested cell pellets were washed once using 1× PBS and lysed with RIPA buffer (Thermo Scientific, Rockford, IL, USA) containing an EDTA-free protease inhibitor cocktail (Roche, Mannheim, Germany) for 1 h at 4 °C. For nuclear and cytoplasmic protein extractions, a NE-PER kit (Thermo Scientific, Rockford, IL, USA) was used. The respective extractions were performed according to the manufacturer’s instructions. An EDTA-free protease inhibitor cocktail (Roche, Mannheim, Germany) was added to CER I and NER buffers prior to the extraction procedure. All extraction steps were carried out at 4 °C or on ice. Fifty micrograms of each protein sample were boiled, electrophoresed and electrotransferred onto polyvinylidene difluoride membranes. These membranes were probed with the following antibodies: anti-IkBα, anti-p38 MAPK, anti-phospho-p38 MAPK (p-p38 MAPK), anti-PARP1 and anti-cleaved PARP1 (all from Cell Signaling Technology, Danvers, MA, USA). Antibody against the NP protein of NDV was a gift from Prof. Khatijah Yusoff and its immunodetection was performed as previously reported [11]. The anti-NF-κB (p65 subunit) antibody used was purchased from Epitomics (Burlingame, CA, USA) while the anti-β-actin antibody was purchased from Sigma Aldrich (St. Louis, MO, USA). Antibody dilutions used were as recommended by the manufacturers. The blots were developed and the resulting proteins bands were quantified as reported previously [5].

Statistical analyses

Quantitative data were analyzed using the GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA, USA) software package. Results were expressed as mean ± standard error of the mean (SEM). The statistical significance of experimental data was calculated using the Student’s t-test throughout this study. A p value <0.05 was considered statistically significant.

Results

NDV induces NF-κB activation through IκBα degradation

NDV infection of 786-O cells led to a reduction in IκBα protein level compared to uninfected cells (Fig. 1a). This reduction appeared to be directly proportional to the multiplicity of infection (MOI) used (Fig. 1a and b), inferring a correlation between NDV infection level and IκBα degradation. Based on a band intensity comparison with the uninfected sample, the reduction was found to be statistically significant, beginning at 1.0 MOI (Fig. 1b). Therefore, further studies were performed at this MOI. To investigate a putative correlation between IκBα degradation and an increase in NF-κB activity, a dual luciferase NF-κB reporter assay was performed. A known reversible inducer of NF-κB, i.e., bacterial lipopolysaccharide (LPS; [12]), was included as a positive control. As expected, LPS induced an activation of the exogenously introduced NF-κB reporter construct in the cells (Fig. 1c). This activation could be suppressed by JSH-23, which is a drug that specifically inhibits LPS-induced NF-κB activation [13]. Cells infected with NDV also showed an increase in their NF-κB activity. Subsequent addition of JSH-23 suppressed this activation, albeit at a minimal level. NDV infection in the presence of LPS resulted in a two-fold increase in NF-κB activity, beyond the levels of their individual activation. Addition of JSH-23 resulted in a suppression of this activity as well.

Fig. 1.

Confirmation of NDV infection, IκBα degradation and NF-κB activation in 786-O cells. a NDV and IκBα proteins detected by specific antibodies in cell lysates of the infected and control cells. b IκBα band intensity in relation to β-actin internal control. Values are presented as means (±SD) of three individual blots. c NF-κB activity measurement using a dual luciferase NF-κB reporter assay. C, mock-infected control. LPS, lipopolysaccharide. *p < 0.05, **p < 0.01, ***p < 0.001

These data are consistent with the notion that the increase in NF-κB activity in the NDV-infected 786-O cells was likely to be due to the degradation of IκBα inside the cells (Fig. 1a). Since NF-κB has been found to be involved in most of the cells’ antiviral responses, particularly at early stages of infection [14, 15], we next set out to examine the kinetics of NDV-induced NF-κB activation in the 786-O cells.

NF-κB activation is associated with its nuclear translocation

NF-κB is known to play a role in mediating immediate early responses to viral pathogens [16]. Its activation is very rapid and does not require de novo protein synthesis [15]. In our preceding experiment (3.1), samples were harvested at 25 h post-infection (hpi). To investigate the levels of NF-κB activation at earlier time points, we performed a kinetic study to monitor NF-κB activity after NDV infection. Measurements were done at 1–3 h intervals post-infection. By doing so, we found that NDV infection significantly (p < 0.05) activated NF-κB as early as 4 hpi (Fig. 2a, solid line). The NF-κB activity increased, commensurate with the rate of infection, reaching a plateau at 13 hpi. The level then dropped at 25 hpi. In an earlier study we observed that, at this time point post-infection, cell death was occurring [5]. Since NF-κB activation is required for IFN-β production [17, 18], we next asked whether IFN-β levels in the cells were affected by NF-κB activation. At 4 hpi, when the NF-κB activity was first increased, a low level of IFN-β was detected in the sample (Fig. 2a, dotted line). This level was not changed at 7 hpi, but was found to be markedly increased at 10 hpi. Similar to the pattern of NF-κB activation, a peak level of IFN-β was seen at 13 hpi, after which the level dropped again at 25 hpi. This latter drop was probably due to viral-mediated cell death [5]. Since the experiment was ended at 25 hpi, it is not known whether the IFN-β level after 25 hpi continues to follow the biphasic level of NF-κB activation. Nevertheless, the gradual increase in IFN-β level subsequent to the increase in NF-κB activity is suggestive of a cause-effect relationship between the two regulatory proteins. Such a scenario is plausible since NF-κB has been shown to be able to bind to the promoter region of the IFN-β gene and to regulate its transcription [18, 19].

Fig. 2.

Activation of NF-κB is associated with its nuclear translocation in infected 786-O cells. a NF-κB activity and Interferon-β level in the infected and control cells. b Detection of NF-κB, IκBα and phosphorylated IκBα (p-IκBα) in the cytoplasmic and nuclear fractions of infected and mock-infected 786-O cells. Values below the blots indicate fold differences of band intensities compared to controls. Each band was normalized to the respective β-actin loading controls. C, mock-infected control

The increase in NF-κB activity in NDV-infected cells at 4 hpi appears to correlate with its nuclear translocation (Fig. 2b). At this 4 h time point, the nuclear NFκB protein level was found to be dramatically increased compared to the uninfected cells. This increase started immediately after the virus adsorption period, which was at 1 hpi. Even though we observed a significant nuclear translocation of NF-κB, its total level in the cytoplasmic fractions did not significantly change compared to the uninfected cells. The reason for this apparent lack of reduction in cytoplasmic NF-κB level remains to be elucidated. Since NF-κB nuclear translocation is known to be due to IκBα degradation [16], it was expected that the cytoplasmic IκBα levels would correspondingly be reduced. In agreement with this notion, cytoplasmic IκBα levels in the infected cells were all reduced compared to the uninfected cells. IκBα requires phophorylation for its degradation [16, 20]. Accordingly, its phosphorylated form was found to be elevated, especially at 2 hpi. Significant changes in the levels of either unphosphorylated or phosphorylated IκBα were not observed in the nuclear fractions of any of the samples. This observation is in agreement with previous reports of a cytoplasmic steady-state localization of IκBα [21, 22].

NDV induces p38 MAPK phosphorylation upstream of NF-κB activation

NF-κB activation in NDV-infected 786-O cells (Fig. 2a) appeared to fluctuate throughout the infection period tested. A similar fluctuating pattern of nuclear NF-κB levels was not readily apparent in the NDV-infected cells (Fig. 3a). The initial increase in nuclear NF-κB activity at 4 and 7 hpi (Fig. 2a) coincided, however, with an increase in cytoplasmic p-IκBα level (Fig. 3a), whereas a second increase in its activity at 13 and 25 hpi (Fig. 2a) correlated with a reduction in cytoplasmic IκBα level (Fig. 3a). A fluctuation pattern of pro-inflammatory cytokines has previously been associated with a bi-phasic phosphorylation of the p38 mitogen-activated protein kinase (p38 MAPK; [23]). The p38 MAPK pathway is known to regulate the transcriptional activity of the NF-κB gene [24, 25]. In the present study, we detected an increase in p38 MAPK phosphorylation as early as 1 hpi (Fig. 3b). At this time point, no NF-κB activity was detected yet (Fig. 2a), despite its nuclear translocation (Fig. 2b). These results suggest that the NDV infection is associated with an induction of p38 MAPK phosphorylation during the 1-h virus adsorption period at the MOI used. The initial phosphorylation that was seen at 1 hpi was reduced at 4 hpi, but increased again at 7 and 10 hpi, followed by another reduction at 13 hpi (Fig. 3b and c). This fluctuating pattern of p38 MAPK phosphorylation and NF-κB nuclear translocation did not affect the intracellular NDV protein accumulation (Fig. 3b). From 4 hpi onwards, the expression of viral proteins continuously increased, reaching a maximum level by the end of the experiment. Infection experiments performed in the presence of the p38 MAPK inhibitors SB202190 and TAK715 revealed a reduction in nuclear translocation of NF-κB in the infected cells (Fig. 3d). The cytotoxicity of the infected cell cultures was also found to be reduced throughout the infection period (Fig.3e).

Fig. 3.

NDV-induced NF-κB activation is preceded by p38 MAPK phosphorylation. a Detection of NF-κB, IκBα and p-IκBα in the cytoplasmic and nuclear fractions of infected and mock-infected 786-O cells. b Levels of p38 MAPK and phosphorylated p38 MAPK (p-p38 MAPK) in total cell lysates of NDV-infected and control 786-O cells. c Relative intensity of p-p38 MAPK in relation to total p38 MAPK. Values are presented as means (±SD) of three individual blots. d Effect of p38 MAPK inhibitors SB202190 and TAK715 on nuclear localization of NF-κB. e Viability of NDV-infected cells in the absence (white bars) or presence (shaded bars) of SB202190 and TAK715. C, mock-infected control

p38 MAPK is phosphorylated in cancer cell lines infected with NDV

786-O is a human ccRCC cell line that is defective for the VHL tumor suppressor gene [26]. To investigate whether a restoration of VHL function in 786-O cells would interfere with the observed NDV regulation of NF-κB and p38 MAPK, we repeated the above experiments with the 786-VHL cell line, in which the VHL function is restored [5]. Similar to 786-O cells, NDV infection increased the level of NF-κB activity in the 786-VHL cells (Fig. 4a). It is interesting to note that this increase was almost double that observed in 786-O cells, despite 786-VHL displaying a significantly lower basal level of NF-κB activity in the mock-infected cells. Previously, it was shown that the level of VHL expression positively correlates with the p38 MAPK activity [27]. In the present study, no NF-κB nuclear translocation was observed (Fig. 4b) even though the level of p-p38 MAPK was increased in the uninfected 786-VHL cells (Fig. 4c). This lack of nuclear translocation is in line with the absence of NF-κB activity in these cells (Fig. 4a). This phenomenon is currently being investigated in further detail. Nevertheless, the infection data confirm that restoration of VHL function in 786-O cells does not interfere with the up-regulation of NF-κB by NDV infection. The increase in NF-κB activity was found to be associated with nuclear translocation of NF-κB (Fig. 4b) and a concomitant increase in p38 MAPK phosphorylation (Fig. 4c). Increased levels of p38 MAPK phosphorylation were also seen in MCF7 (human mammary gland adenocarcinoma) and Saos-2 (human osteosarcoma) cells infected with NDV (Fig. 4d), indicating that this is a general phenomenon occurring in unrelated cancer cell types.

Fig. 4.

p38 MAPK is phosphorylated in cancer cell lines infected with NDV. 786-O, 786-VHL, MCF7 and Saos-2 cell lines were infected with NDV. a Levels of NF-κB activity and b NF-κB nuclear translocation examined in 786-O and 786-VHL cells. c Levels of p38 MAPK and p-p38 MAPK. d Comparison of p-p38 MAPK levels in 786-O, 786-VHL, MCF7 and Saos-2 cells. Data are presented as fold difference of band intensities compared to controls. Each band was normalized to its respective β-actin loading control. Values are presented as means (±SD) of three individual blots

NDV-induced NF-κB activation correlates with PARP cleavage and apoptosis

NF-κB is known to be involved in both cell survival and apoptosis [28]. Recently, we have shown that NDV induces apoptosis in 786-O cells 24 hpi [5]. This apoptosis was confirmed via terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) as well as DNA fragmentation assays. A similar TUNEL result was seen in the present experiments in cells harvested at 25 hpi (Fig. 5d). Cleavage of the poly-(ADP-ribose) polymerase (PARP) is an established indicator of caspase-mediated apoptosis [29]. In the present study, NDV infection resulted in the appearance of cleaved PARP1 in the infected cells from 10 hpi onwards (Fig. 5a). The amount of cleaved PARP1 increased with infection time and peaked at 25 hpi. This PARP1 degradation correlated with a decreasing viability of the infected cells (Fig. 5b). A kinetic cell viability study showed that almost all of the cells were dead after 73 h of infection. The morphology of the dead cells closely resembled that of cells undergoing apoptosis, i.e., plasma membrane blebbing and detachment from the culture flask surface (Fig. 5c).

Fig. 5.

NDV infection leads to PARP1 cleavage and cell death through apoptosis. a Total PARP1 levels did not change significantly in all of the infected and control samples. Cleaved PARP1 was first seen at 10 hpi. b Reduction of cell viability in NDV-infected 786-O cells. c Morphological changes in NDV-infected cells, including membrane blebbing and cell detachment from tissue culture surfaces. d Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) analysis of NDV-infected 786-O cells after 25 hpi. Cells were counterstained with propidium iodide (PI)

Discussion

Our results show that NDV infection causes a multiplicity of infection (MOI)-dependent degradation of IκBα in 786-O clear cell renal cell carcinoma (ccRCC) cells. Degradation of the IκBα protein is known to lead to activation of NF-κB [16]. This activation was also seen in the present study. Interestingly, an additive effect of NF-κB was seen in LPS-treated cells co-infected with NDV. This additive effect suggests the involvement of separate mechanistic pathways in NF-κB activation. This notion is further supported by a minimal NF-κB inhibition observed in NDV-infected cells in the presence of JSH-23. These results infer that NDV-induced NF-κB activation is not completely inhibited by JSH-23, at least in 786-O cells and at the MOI used in this study. It is known that JSH-23 inhibits the NF-κB canonical pathway, but not the non-canonical pathway [30]. Crosstalk between the canonical and non-canonical NF-κB pathways has been shown to be important in the regulation of antiviral innate immunity [31, 32]. Thus, it is likely that a similar response also occurs during NDV infection. Currently, we are performing additional studies to test whether this is indeed the case.

NDV-induced NF-κB activation in 786-O cells was detected as early as 4 hpi and was found to peak at 13 hpi. This activation was followed by appearance of IFN-β in the culture medium. The requirement for NF-κB activation in IFN-β production has previously been reported [17, 18]. In the present study, the IFN-β secretion pattern appeared to follow the NF-κB activation pattern, suggesting a possible correlation in the response of 786-O cells to NDV infection. Previously, we found that the level of NDV-induced IFN-β secretion in 786-O cells was almost 3-fold lower than that in 786-VHL cells [5]. Accordingly, in the present study, NF-κB activity in NDV-infected 786-O cells was nearly 3-fold lower than that in 786-VHL cells. Taken together, our results indicate that NDV infection causes a more robust NF-κB activation and concomitant IFN-β secretion in 786-VHL cells compared to 786-O cells. This activation leads to a higher killing effect in 786-VHL cells, despite the fact that loss of VHL activity is known to activate NF-κB [33], a condition seen in the mock-infected control cells.

Inactive NF-κB is known to bind to inhibitors of κB (IκB) and to be sequestered in the cytoplasm [28]. Specific stimuli cause phosphorylation and subsequent degradation of IκBα. This degradation leads to the release and translocation of NF-κB to the nucleus and, at this site, to the up-regulation of its target genes [28]. In NDV-infected 786-O cells co-transfected with pGL4.32 the NF-κB luciferase signal, which was increased at 4 hpi, also correlated with NF-κB nuclear translocation and IκBα degradation. The phosphorylated form of IκBα is known to be subject to degradation via the ubiquitin-proteasome pathway [28]. Even though pIκBα appeared to be slightly decreased at 1 hpi, this decrease was not statistically significant. At 2 hpi, a significant increase in its level was found to correlate with the nuclear translocation of NF-κB. Interestingly, throughout the NDV infection period, a fluctuating pattern of NF-κB activation and a marginally similar pattern of nuclear translocation was observed in 786-O cells. A similar fluctuation was also previously seen after infection of cells with other viruses, such as human cytomegalovirus [34] and feline infectious peritonitis virus [23]. It has been suggested that this type of signal fluctuation may be associated with different stages of viral infection [35].

Previously, Hirasawa et al. proposed that p38 MAPK may be involved in translation of the encephalomyocarditis virus RNA [36]. In the present study, the p38 MAPK and NF-κB signal fluctuations did not significantly affect the overall intracellular level of viral protein accumulation. Infection experiments in the presence of the p38 MAPK inhibitors SB202190 and TAK715 reduced the nuclear translocation of NF-κB which, in turn, correlated with a reduction in NDV-induced cytotoxicity. These data suggest a link between p38 MAPK/NF-κB activity and the NDV-mediated killing effect. The exact mechanism underlying this p38 MAPK/NF-κB phosphorylation pattern in NDV replication, however, requires further investigation. Nonetheless, our study shows that p38 MAPK is one of the earliest cellular responses to be activated following NDV infection.

Mitogen-activated protein kinase (MAPK) signaling involves three major pathways. The first one involves the extracellular signal-regulated kinases 1 and 2 (ERK1/2), mediators of proliferative stimuli [37]. The second one involves the c-Jun N-terminal kinases (JNKs), mediators of extracellular stress responses [37]. The third one involves p38 MAPK, which is also involved in mediating extracellular stress responses, particularly through regulating the expression of cytokines [38]. Recently, Bian et al. reported that p38 MAPK is activated in NDV-infected cells [39]. They detected phosphorylation of p38 MAPK as early as 4 hpi using an MOI of 10. They did not, however, look at earlier time points. In the present study, p38 MAPK phosphorylation was detected immediately following the virus adsorption period, which was designated as 1 hpi. At this time point, no increase in NF-κB luciferase reporter activity was observed, even though NF-κB was already detected in the nuclear fractions. Additional experiments in the presence p38 MAPK inhibitors affected the nuclear localization of NF-κB in NDV-infected cells at later time points. These data indicate that p38 MAPK activation occurs upstream of NF-κB activation in NDV-infected 786-O cells. Thus, it is conceivable that phosphorylation of p38 MAPK at 1 hpi may have resulted in IκBα degradation which, subsequently, may have led to NF-κB translocation to the nucleus. A schematic overview of the putative pathways involved, including STAT, is shown in Fig. 6.

Fig. 6.

A schematic representation of the putative signaling pathways involved in NDV-mediated apoptotic death in infected cancer cells

Even though NF-κB activation has been associated with pro-survival pathways [40], it has also been associated with apoptosis induction [41]. In the present study, the detection of PARP1 degradation following NDV infection provides a possible link between NDV-mediated NF-κB activation and apoptosis in the infected 786-O cells. PARP plays an important role in DNA repair processes and programmed cell death (apoptosis) [42]. Therefore, detection of proteolytically cleaved PARP is considered a reliable marker of apoptosis [43]. We observed that PARP cleavage in infected 786-O cells was followed by a reduction in cell viability, culminating in almost 100 % cell death at 73 hpi.

Overall, our study suggests that NDV is capable of inducing activation of the p38 MAPK/NF-κB/IκBα pathway in ccRCC cells, which results in cell death due to apoptosis. Additional information regarding this pathway, and the exclusive IFN-β induction by NDV [5], may contribute to improving strategies to use NDV as an oncolytic agent in clinical settings. This is possible, given the central role of NF-κB in the type I IFN antiviral innate immune response, which can be manipulated to facilitate clearance of damaged and cancerous cells. Preliminary data on two other cancer-derived cell lines (MCF-7 and Saos-2) also showed increases in p38 MAPK phosphorylation following NDV infection. This novel finding provides important information regarding the types of anti-viral responses that can be generated by NDV-infected cancer cells. Mechanistic insights into the responses of cancer cells to NDV virus infection, as presented here, will help to unravel the full potential of this virus as a potent oncolytic therapeutic agent in cancer treatment.