This study was conducted to characterize the combined effects of irradiation and HSV-1 infection on an immortal oral epithelial (HMK) cell line. The most important result was that after 144 hours in culture, irradiation lead to increase in the viability of this immortal cell line and the effect was potentiated by HSV-1 infection in cells infected with a low MOI. Simultaneously, the expression of caspases 3, 8, and 9 was downregulated in HSV-1 infected and irradiated cells, but bcl-2 was upregulated. This cannot be due to the overall general shutoff of gene expression caused by progressive HSV-1 infection, as nuclear factor κB was significantly upregulated in this time point compared to the nonirradiated infected cells. This primary observation suggests that HSV-1 infection and irradiation, to both of which oral epithelial cells are frequently exposed, might aid transformed cells to resist the toxic effects of HSV-1 infection and even gain a viability advantage. Downregulation of caspases essential in apoptosis might be one of the pathways involved in this effect. Although this is still too short a time to establish conclusions with regard to in vivo tumorigenesis, these in vitro results warrant further studies because they can be clinically important. The gingival tissue around the teeth is frequently exposed to ionizing radiation during dental treatments and the effect can be even potentiated by the scattered irradiation due to metallic implants or fillings . Also, reactivation of HSV-1 infection and viral shedding in saliva is frequent. Even more important is that the presence of HSV-1 in HNSCCs might affect patient survival after treatment with surgery with radiotherapy or chemoradiotherapy .
The assay used to study the cell viability is based on ATP detection that correlates with the number of living cells [25, 26]. HSV can enhance the glycolysis of the infected cell but this effect is not a significant source of upward bias in these experiments, as demonstrated by data from Peri et al. where uninfected cells were compared to several HSV-1 mutant- and wild-type strains . Also, ATP is quickly degraded outside the cell and therefore already dead cells cannot increase the “viability” ratings of the culture analyzed.
Immortal cell lines
In the present study HSV-1 infection and irradiation had near-opposite effects on cell viability of HaCaT cells and HMK cells. This can be partly explained by the origin of these two cell lines, skin (HaCaT) and oral mucosa (HMK), and/or the differences in their genome. Although both are spontaneously immortalized nontumorigenic keratinocyte cell lines, HMK cells could be considered more abnormal than HaCaT cells, presenting a totally tetraploid karyotype (DNA index 2.01). Spontanously immortalized HaCaT cells were found to be hypotetraploid with a DNA index of 1.92 [27, 28]. It is impossible to obtain karyotypically similar two cell lines However, there are no earlier comparative studies on radioresistance, permissivity to virus infections or cell viability done with different spontaneously immortal cell lines. Although genetically quite different, these two cell lines support similar rates of HSV-1 infection as shown here although the replication starts slower in the HMK cells.
One of the limitations of this study is that the cells were around 80% confluent at the time of infection (or mock infection) and continued to increase in number until past the 24 hours post-irradiation time point. Caspase and NFκB1 expression was found to be increased in the uninfected and nonirradiated HMK cells at 72 hours. There is a possibility that supraconfluency of the cultures was reached somewhere between the 24 hours and 72 hours, which might induce differentiation-related signals to which caspases also belong in human keratinocytes . Apparently, these effects have dissipated by the 144 hour time point since the caspase levels in uninfected and nonirradiated cells return to the 24-hour levels. Also, this effect was not seen in irradiated cells but as irradiation has profound effects on differentiation-related signaling, the effect of confluency on 72 hour nonirradiated cultures remains likely .
The effect of irradiation
Transcriptional activation of cell death regulatory genes is of the utmost importance for cellular radiosensitivity . BcI-2 has been shown to protect cells from irradiation-induced cell death . On the contrary, NFκB1 is activated by the ATM kinase following irradiation  and confers resistance to apoptosis which can be abrogated by blocking NFκB1, leading to cytotoxicity and caspase 3 activation in cancer cell lines after irradiation . Irradiation affected the uninfected HMK cells by inducing NFκB1 and Bcl-2 expression at 24 hours but no effect was seen on caspase expressions. The Bcl-2 upregulation continued up to 72 hours when NFκB1 and caspase 9 and 3 expressions declined. The effects of irradiation on uninfected cells’ gene expression mainly dissipated by 144 hours. By this time, however, increased cell viability was found in irradiated cells, which reflect an increased resistance to radiation-induced damage. This could be explained by earlier upregulation of the antiapoptotic genes and lower activity of the mitochondrial apoptotic pathway indicated by the lower expression levels of caspases 9 and 3 at 72 hours and increase in the Bcl-2 expression from 24 to 72 hours. We found also that NFκB1 was upregulated 24 hours postirradiation but returned back to baseline level at 144 hours. It might be that upregulation of NFκB1 found at 24 hours is due to the genotoxic stress which allows DNA damage repair and cell survival as shown by Janssens et al., .
The effects of HSV-1
A lytic HSV-1 infection almost always destroys its host cell. However, there is evidence that HSV-1 may latently or nonproductively infect the epithelial cells as well . Accordingly, HSV-1 infection together with other cofactors like irradiation might cause changes in spontaneously immortal cells toward malignancy. HSV-1 infection has been found to activate the NFκB1 transcription factor to prevent the target cell from undergoing apoptosis . Interestingly, our results pointed out that HSV-1 elevated Bcl-2 and NFκB1 expression already at 24 hours after mock-irradiation but there were not yet any effects on cell viability. As the infection progressed, caspases 3, 8 and 9 were downregulated compared to the uninfected cultures at 72 hours, surprisingly together with NFκB1, but the Bcl-2 levels were increased. This may be due to the accelerating spread of HSV-1 in the cultures at this time point, leading first into a well-known evasion of apoptosis caused by the expression of typical HSV-1 antiapoptotic proteins gD, gJ, Us3, ICP27 and ICP4 until the infection nears its completion at 144 hours [8, 19–21]. The lowest MOI causes more robust changes in NFκB1 and Bcl-2 levels as demonstrated by the lack of statistical significance in the changes in expression levels found with higher-MOI, although the higher-MOI expression levels displayed a trend in the same direction. Aubert et al. implied that HSV-1 blocks apoptosis by targeting Bax and thus preventing mitochondrial cytochrome c release and therefore activation of caspase 9 in human epithelial cells . Bcl-2 can act as an inhibitor of Bax [31, 37, 38]. Therefore in HSV-1 infected cells, upregulated Bcl-2 might heterodimerize with Bax and block apoptosis, similarly as described earlier in non-irradiated cells .
Our results showed that HSV-1 infection at 144 hours had progressed to a point where most of the cells were nearly completely infected. HSV-1 infection is thought to trigger caspase 9-mediated apoptosis, caspase 3 being important for the downstream apoptotic pathway [13, 16]. It has been suggested that HSV-1 induces apoptosis by first triggering the cytochrome c release from the mitochondria, thus activating caspase 9 leading to apoptosome formation and caspase 3 cleavage . However, our results showed that caspases 8 and 3, but not caspase 9, were upregulated at 144 hours due to HSV-1. This difference may be caused by the differences in the cell lines studied, as our HMK cells are HPV negative, and the epithelial cells used by Aubert et al. are known to be HeLa contaminants containing HPV-18 which affects apoptosis, especially via caspase 8 [39, 40].
The combined effects of HSV-1 and irradiation
The most intriguing aspect of our data stems from the major differences in gene expression and viability responses of the cultured cells when the combined effects of irradiation and HSV-1 infection are compared to the effects of either exposure only. As seen in our results, at 144 hours the HSV-1 infection has spread to most of the cells in culture. Therefore the HSV-1 mediated antiapoptotic effects are best represented in the 144 hour results as the previous time points are less representative of the HSV-related effects. The combined effects of HSV-1 and irradiation did not cause any additional toxicity as determined by the viability assays. Irradiation of HSV-1 infected cells resulted in upregulation of caspase 3, caspase 8 and downregulation of caspase 9 at 24 hours. Simultaneously, NFκB1 was upregulated in all irradiated cultures compared to their nonirradiated counterparts regardless of HSV-1 presence. Therefore the immediate NFκB1 response to radiation does not seem to be affected by HSV-1. At 72 hours, bcl-2 and caspase 3 were upregulated and caspases 8 and 9 downregulated. Interestingly all caspases were downregulated at 144 hours while both NFκB1 and bcl-2 were upregulated. Since ICP27 is important in prevention of apoptosis , it is tempting to speculate whether ICP27 plays a role in the effects found here, partly because the highest ICP27 expression was detected at 144 hours in the irradiated infected cultures. Therefore, the role of ICP27 in irradiation induced apoptosis needs further study. In our experiment, HSV-1 infection was largely unaffected by irradiation as determined by VP16 qRT-PCR, virus culture and staining for HSV-1 gC. This would indicate that at 2 Gy, HSV-1 survives the irradiation and its infection rate remains unaffected.
Recently Dufour et al.  showed that HSV-1 ribonucleotide reductase R1 (rR) protects cells from apoptosis by binding to caspase 8. Spear et al.  reported that infection with rR –defective HSV-1 leads to increased apoptosis as measured by FACS analysis. When their results at 72 hours post irradiation are examined more closely, the tumor cells infected with HSV-1 had double the amount of apoptosis than the cells infected with the same virus but combined with 2 Gy irradiation. 2 Gy irradiation in itself had a negligible effect on apoptosis in their experiment. In the present study, contrary to the results of Spear et al. at the same time point, apoptotic gene expression was not present at high levels and no effects in cell viability were observed before the 144 hour time point, not included in their data.
After 144 hours in culture, cell viability was gradually lowering in nonirradiated infected cultures and apoptotic markers caspase 3 and 8 upregulated together with a decline in bcl-2 due to advanced HSV-1 infection, supporting the current literature on HSV-1 related apoptosis [13–16, 42]. However, the most striking effect observed in this study is that irradiation of HMK cells with 2 Gy with or without HSV-1 infection does not actually lower the viability of the cells or lead to outright cell death during the study period and even lead to an elevation in cell viability. The combined effects exerted a profound abrogation of the expression of all caspases studied while NFκB1, that until 144 hours had remained relatively constant, strongly upregulated implicating the NFκB1 pathway as a mediator of long-term radiation responses in HSV-1 infected cells. NFκB1 has diverse roles in cellular apoptosis  and inhibition NFκB has been linked to apoptosis and delayed cell growth . Therefore its upregulation may have contributed the effects we observed. The NFκB1 pathway, when activated, leads to higher bcl-2 expression and therefore lower expression of apoptotic markers such as caspase 3 . This is clearly supported by our findings in irradiated and infected cells. Bcl-2 is implicated in resistance to radiation therapy and chemotherapeutic agents [32, 46]. Its expression displayed a downward trend in time, but remained at a higher level by the end of the experiment in irradiated cells, particularly those infected with 0.0001 MOI. This implies that bcl-2 may contribute to the observed downregulation of the intrinsic apoptotic pathway.
Because effects on cell viability were also seen using a UV inactivated virus that causes no visible HSV-1 plaque formation, it is possible that these effects might be at least partly mediated by the effects of innate immunity. This would be plausible, given that the tissue wouldn’t need to be completely infected with HSV to have far reaching effects. However, the presence of HSV-1 would still be required. Irradiation induces a wide variety of innate immunity related genes such as TNF-α and IFN-γ . TNF-α has been linked to radioresistance of oral cancer cells whereas IFN-γ is able to induce cathepsin S expression which leads to radioresistance [48, 49]. These factors could contribute to the effects observed in our study.
To summarize, after six days in culture the combined effects of HSV-1 infection and 2 Gy irradiation lead to an increase in NFκB1 and bcl-2 expression, significantly lower expression of caspases 3, 8, and 9 and higher viability ratings as compared to the non-irradiated infected cultures, but also seen using a UV-inactivated virus. Apoptotic pathways are possibly involved in these effects. As oral epithelial cells are coexposed to HSV-1 infection and radiation during radiotherapy or dental radiographic exposures, there might be an increased risk for cellular transformation in subjects who are exposed to other common carcinogens, such as tobacco and alcohol. Future studies are needed to explore the significance of the present results in clinical settings.