Combined effect of the pro-apoptotic rhTRAIL protein and HSV-1 virus in head and neck cancer cell lines

Knowledge on the molecular and clinical characteristics of head and neck squamous cell carcinoma (HNSCC) is vast. However, an effective therapy that increases the life expectancy of these patients, with a 5-year overall survival of 50%, is still unknown. Here we evaluated the combined effect of the pro-apoptotic protein rhTRAIL with the replication-competent wild-type HSV-1 virus in head and neck cancer cell lines. We observed a difference in the modulation profile of proteins related to apoptotic pathways in the studied cell lines. The HCB289 exhibited caspase-9 activation in the presence of the HSV-1 virus, while the UD-SCC-2 exhibited caspase-8 activation in the presence of rhTRAIL. Both cell lines exhibited PARP activation by combining rhTRAIL and HSV-1 virus treatment. Flow cytometry analysis exhibited greater induction of late apoptosis for the HCB289 and UD-SCC-2 after the combination treatment of the HSV-1 and rhTRAIL. However, the UD-SCC-2 also presented induction of late apoptosis by the presence of rhTRAIL in monotherapy. These data suggest an enhancement of the effect of the combination treatment of the rhTRAIL and the HSV-1 on reducing viability and induction of cell death.

www.nature.com/scientificreports/Apo2L/TRAIL triggers apoptosis by binding to functional death receptors (DR4 and DR5) 17,18 , but, regardless of the fact that these receptors are highly expressed in HNSCC lines, many tumor cell lines end up showing resistance to treatment with TRAIL monotherapy 18,19 .
Thus, combining both therapeutic approaches, oncolytic virotherapy and TRAIL-mediated apoptosis, may be promising in for HNSCC treatment.In this work, we evaluated the combined treatment of rhTRAIL protein and replication-competent wild-type HSV-1 virus in HNSCC cell lines.
After 48 h of treatment, cell lines showed a decrease in cell viability when treated with WT HSV-1 at MOI 1 alone or in combination with rhTRAIL.HaCat also showed a decrease in cell viability of 22.47%, when treated with HSV-1 at MOI 0.2 (Fig. 1A).For 72 h, we observed the HCB289, UD-SCC2 and FaDu showed a significant effect upon exposure to the combination of rhTRAIL and WT HSV-1.Although we used the HaCat as a normal control and the rhTRAIL was previously described as promoting apoptosis mainly tumor cells, it was affected by the rhTRAIL treatment, which was not expected.The UM-SCC-47, UM-SCC-104 and HaCaT, also showed a decrease in viability when exposed to the virus alone (Fig. 1B).

HCB289 and UD-SCC-2 has different modulation pattern of proteins associated with the apoptotic pathway
To characterize the possible mechanisms responsible for cell viability decreases in HNSCC cell lines, we investigated the protein expression modulation associated with intrinsic and extrinsic apoptotic pathways after 24 www.nature.com/scientificreports/and 48 h of exposure to rhTRAIL and its combination with the WT HSV-1.We selected the HCB289 and UD-SCC-2 for a cell death function characterization.This selection was made from the observation that for both, 48 and 72 h, these HNSCC cell lines showed a significant reduction in cell viability to exposure to the combination rhTRAIL and virus compared to rhTRAIL or virus as monotherapy.
Results for HCB289 showed modulation on caspase-9 and PARP cleavages after 24 h of exposition with the combination of rhTRAIL and virus, especially under conditions of the virus at MOI 1.0 of 287.63% and 402% relative expression respectively compared to negative control.However, the effect on caspase-8 cleavage was evident only with the rhTRAIL and HSV-1 MOI 1 combination (2.7 times higher than control group).On the other hand, we did not observe a significant change in the caspase-3 cleavage under different treatment conditions (Fig. 2A,B).
For UD-SCC-2, the 24 h exposition treatment promoted alterations on incrementation of caspase-8 cleavage when exposed to rhTRAIL alone (346.92%) or in combination with HSV-1 at MOI 0.2 (286.84%) or 1.0 (324.09%).Besides, PARP cleavage significantly increases at HSV-1 presence at MOI 1.0 (3.19 times higher) and in combination with rhTRAIL on both concentrations (3.65 and 3.77 times higher).Increased Caspase-3 cleavage was reported at rhTRAIL monotherapy (3.38 times higher) and in the combination of HSV-1 at MOI 1.0 (2.45 times higher) compared to negative control.No differences were detected for caspase-9 cleavage under different treatment conditions (Fig. 3A,B).

rhTRAIL and HSV-1 induce apoptosis in HCB289 and UD-SCC-2 cell lines
Apoptosis flow cytometry analysis of rhTRAIL and HSV-1 treatments showed an increase in late apoptosis and necrosis in HCB289, 24 and 48 h post-treatment, mainly when combined with HSV-1 at MOI 1 of 52.12% and 44.62% respectively (Fig. 6A,B).Similarly, combined treatments in the UD-SCC-2 cell line showed an increase in late apoptosis and necrosis, but also the rhTRAIL monotherapy showed an induction of 25.7% and 51.15% for 24 and 48 h post-treatment, respectively (Fig. 6C,D).We also compared the treatment conditions for each cell population (viable, necrotic, early apoptotic and late apoptotic) which demonstrated that the treatment was responsible for decreased viable cells detection as the cells entered the cell death process, mainly for rhTRAIL and HSV-1 presence for HCB289 (from 83.5%, negative control, to 47.4%, rhTRAIL + HSV-1 MOI 1.0 for 24 h posttreatment) and rhTRAIL alone (41.8% for 48 h) and in combination with the virus for UD-SCC-2 (42.3% at MOI 1.0 for 48 h posttreatment) (Supplementary Fig. S1).
The dot plots graphs generated by flow cytometer sample readings demonstrated a greater presence of double markers (annexin V-PE and 7AAD) by combining rhTRAIL and HSV-1 at MOI 1.0 for both HCB289 (50.9% for 24 h) and UD-SCC-2 (47.7% for 48 h) cell lines (Supplementary Figs.S2-S5).

HSV-1 does not modulate the DR-5 receptor expression
Death receptor 5 (DR-5) expression was examined to investigate the possible mechanism responsible for modulating the apoptotic pathway triggered by the virus and the rhTRAIL ligand.After 24 and 48 h of the treatment with rhTRAIL, HSV-1 or both, no differences are observed in DR-5 receptor expression on HCB28 and UD-SCC-2 cell lines (Figs. 7, 8).These results suggest that other cellular mechanisms may be responsible for the biological effects reported in the present study.

Discussion
This study explores an advanced head and neck cancer treatment using wild-type HSV-1 infection in combination with the rhTRAIL protein.TRAIL ligand can be naturally released as a soluble form by proteolysis, mostly promoting apoptosis in tumor cells 25,26 .The soluble form of TRAIL is well tolerated by non-cancer cells; and traces can be found in the healthy adult plasma (~ 100 pg/mL) 27,28 .Moreover, its cognate functional cell death receptors, DR4 and DR5, are ubiquitous expressed in several tumor cell types 17 .Several clinical studies are based on recombinant human TRAIL protein (rhTRAIL) and agonist antibodies to TRAIL receptors.However, none of these molecules alone or combined with other therapeutic agents have acquired the efficacy expected in clinical trials 20,25 .www.nature.com/scientificreports/Currently, many therapies use gene therapy to target the TRAIL gene and combinatorial use of this ligand with various other molecules, such as chemotherapies 24,[29][30][31][32] .Griffith and colleagues showed, using a TRAILexpressing adenovirus, that the rapid expression of TRAIL protein in tumor cells triggered apoptosis via caspase-8 activation 29 .A similar response was observed in our study, where rhTRAIL in monotherapy or combination with HSV-1 virus induced apoptosis by activating caspase-8 cleavage in the UD-SCC-2 cell line.
Jeong and Yoo treated 14 tumor cell lines and a colorectal cancer murine model with a TRAIL/ANGPT1-coexpressing oncolytic vaccinia virus, showing that the dual expression of these molecules was effective in inducing apoptosis and inhibiting tumor growth in the in vivo assays, followed by caspase 3 and CD8 expression in tumors 30 .Likewise, we showed by flow cytometry analysis that the combination of HSV-1 virus and rhTRAIL treatments induced apoptosis in the HCB289 and UD-SSC-2 cell lines .Besides, TRAIL-treated UD-SCC-2 cell line also showed induction of late apoptosis/necrosis as monotherapy.These results demonstrate the potential of combining therapeutic approaches to attack tumor cells.
In the same way, Gao and colleagues demonstrated a synergistic effect when treating colorectal cancer cell lines with a TRAIL-expressing oncolytic adenovirus in combination with radiotherapy 31 .Authors showed that the armed oncolytic virus triggered apoptosis through caspase-3 and caspase-8 cleavage in the SW480 cell line, mainly when combined with radiotherapy 31 .In our work, we showed apoptosis involvement by PARP cleavage when HCB289 and UD-SCC-2 cell lines were treated with the combination of rhTRAIL and HSV-1 virus.
Moreover, as shown by Hu and co-workers, when treating lung tumor cell lines with a poxvirus-derived oncolytic virus expressing TRAIL, cells undergone apoptosis, while the parental poxvirus non-expressing TRAIL triggered RIP3-dependent necrosis 32 .Although in our study, we did not evaluate the expression of RIP1/RIP3, we observed cell-and treatment-specific cell death response profiles.Thus, HSV-1 virus induced caspase-9 cleavage in the HCB289 cell line, while caspase-8 cleavage was observed when rhTRAIL was inoculated, either alone or in combination with the virus.In another study, Kaoru and colleagues tested in glioblastoma (GBM) and stem cell-derived (GSC) cell lines the biological effect of an HSV-1-based oncolytic virus (oHSV), TRAIL, and a recombinant virus with TRAIL expression (oHSV-TRAIL) 24 .They observed by western blotting that cleavage of PARP was not detected by either TRAIL or oHSV as monotherapy in glioblastoma cells, indicating that oHSV-mediated cell death is primarily not dependent on caspase/PARP activation 24 .In contrast, oHSV-TRAIL showed cleavage of caspase-8, -9, and PARP and induction of apoptosis in the cell lines resistant to TRAIL or oHSV alone 24 .The authors inferred that oncolytic virus-based gene therapies might be feasible for different tumor types with different molecular profiles 24 .Our study also did not detect PARP cleavage by TRAIL monotherapy, but it was possible to detect PARP cleavage for the HCB289 cell line in the presence of HSV-1 virus at both concentrations (MOI 0.2 and MOI 1.0).This highlights that the HCB289 cell line resistant to TRAIL monotherapy could be induced apoptosis by the presence of the virus.
Although rhTRAIL was described as able to induce apoptosis mainly on tumor cells 15 , we showed that the HaCaT cell line also showed a reduction in cell viability.Although often used as a normal control, it is worth mentioning that HaCaT is an aneuploid immortalized keratinocyte cell line, thus a limitation of our work.Regardless of this fact, the main objective of this proof-of-concept work was to evaluate the combined effect of rhTRAIL and HSV-1 oncolysis against HNSCC.In this regard, authors understand that the use of a wild-type HSV-1 virus could never be used in a clinical trial setup, and we point this out as a limitation of the study.To this end, additional experiments must be conducted to further establish proper attenuation of the therapeutic approach.
To our knowledge, this is the first study combining HSV-1 virus and TRAIL as therapy for HNSCCs, showing that the combination potentiates the viability reduction through cell death induction.Our results open new therapeutic approaches for HNSCC patients, which should be explored in further studies.

Cell lines
Vero cells (ATCC #CCL-81, Manassas, VA, USA) were used to produce the wild-type HSV-1 virus.Four HNSCCderived cell lines (UD-SCC-2, gently donated from Henning Bier; UM-SCC-47, Merck Millipore, Burlington, Genotyping confirmed the identity of all cell lines, as previously described by Oliveira et al 33 .Moreover, all cell lines were tested for mycoplasma through the MycoAlert™ Mycoplasma Detection Kit (Lonza Bioscience, catalog no.LT07-318, Morrisville, NC, USA), following the manufacturer's instructions.

HSV-1 viral production
The virus used in this work was an HSV-1 virus (strain 17), cloned as a bacterial artificial chromosome 34 (gently donated from David A. Leib, Geisel School of Medicine at Dartmouth, Department of Microbiology and Immunology).Viral amplification was performed at MOI 0.01 in Vero cells in M199 medium supplemented with 1% FBS (Thermo Fisher Scientific, Waltham, MA, USA) and 1% penicillin/streptomycin (Sigma-Aldrich, San Luis, MO, USA), in a humidified incubator at 37 °C with 5% CO 2 .Viral particles were collected 72 h postinfection.Intracellular viral particles were extracted by three cycles of freeze and thaw, followed by three cycles of sonication.Cellular debris were eliminated by centrifugation at 600 × g for 20 min (4 °C), and supernatants were collected and stored at -80 °C until required.

Figure 1 .
Figure 1.rhTRAIL exposure effect alone and in combination with WT HSV-1 virus on HNSCC cell lines and the normal cell line, HaCaT.Treatment was performed with rhTRAIL at 100 ng/mL and in combination with WT HSV-1 virus at MOI 0f.0.2 and 1.0.Cell viability analysis after 48 (A) and 72 h (B) of exposure.Results are expressed by mean percentage ± standard deviation of cell viability relative to PBS 1X (considering 100% viability).Data represent the average of three independent assays performed in triplicate.The asterisks indicate statistical significance (*P < 0.05; **P < 0.01; ***P < 0.001) in the comparison with the experimental groups by non-normal distribution Kruskal-Wallis test.

Figure 2 .
Figure 2. Expression analysis of apoptosis-associated proteins in the HCB289 cell line after 24 h of exposure with rhTRAIL ligand, WT HSV-1 and in combination.(A) and (B) detection and quantification of proteins after treatment.Results are expressed as mean percentage ± standard deviation of relative expression of cleaved protein relative to control (considering 100% expression) and normalized by the portion of total protein (noncleaved).Data represent the average of three independent assays.The asterisks indicate statistical significance (*P < 0.05; **P < 0.01; ***P < 0.001) in the comparison with the experimental groups by one-way ANOVA.Raw blots are presented in Supplementary Data 1.

Figure 3 .
Figure 3. Expression analysis of apoptosis-associated proteins in the UD-SCC-2 cell line after 24 h of exposure with rhTRAIL ligand, WT HSV-1 and in combination.(A) and (B) detection and quantification of proteins after treatment.Results are expressed as mean percentage ± standard deviation of relative expression of cleaved protein relative to control (considering 100% expression) and normalized by the portion of total protein (noncleaved).Data represent the average of three independent assays.The asterisks indicate statistical significance (*P < 0.05; **P < 0.01; ***P < 0.001) in the comparison with the experimental groups by one-way ANOVA.Raw blots are presented in Supplementary Data 2.

Figure 4 .
Figure 4. Expression analysis of apoptosis-associated proteins in the HCB289 cell line after 48 h of exposure with rhTRAIL ligand, WT HSV-1 and in combination.(A) and (B) detection and quantification of proteins after treatment.Results are expressed as mean percentage ± standard deviation of relative expression of cleaved protein relative to control (considering 100% expression) and normalized by the portion of total protein (noncleaved).Data represent the average of two independent assays.The asterisks indicate statistical significance (*P < 0.05) in the comparison with the experimental groups by one-way ANOVA.Raw blots are presented in Supplementary Data 3.

Figure 5 .
Figure 5. Expression analysis of apoptosis-associated proteins in the UD-SCC-2 cell line after 48 h of exposure with rhTRAIL ligand, WT HSV-1 and in combination.(A) and (B) detection and quantification of proteins after treatment.Results are expressed as mean percentage ± standard deviation of relative expression of cleaved protein relative to control (considering 100% expression) and normalized by the portion of total protein (noncleaved).Data represent the average of two independent assays.The asterisks indicate statistical significance (*P < 0.05) in the comparison with the experimental groups by one-way ANOVA.Raw blots are presented in Supplementary Data 4.

Figure 6 .
Figure 6.Flow cytometry apoptosis analysis after treatment with rhTRAIL ligand, WT HSV-1 and in combination.A and B) Analysis of HCB289 cell line after 24 and 48 h of treatment exposure.C and D) Analysis of UD-SCC-2 cell line after 24 and 48 h of treatment exposure.Results are expressed as mean percentage ± standard deviation of the cell death rate.Data represent the average of three independent assays.The asterisks indicate statistical significance (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.001) in the comparison with the experimental groups by one-way ANOVA.

Figure 8 .
Figure 8. DR-5 expression analysis in the UD-SCC-2 cell line after treatment with rhTRAIL ligand, WT-HSV-1 and combination.(A) and (B) Detection and quantification of the receptor after 24 h of exposure treatment.(C) and (D) Detection and quantification of the receptor after 48 h of exposure treatment.Results are expressed as mean percentage ± standard deviation of relative expression of cleaved protein relative to control (considering 100% expression) and normalized by the protein α-tubulin.Data represent the average of two independent assays.Raw blots are presented in Supplementary Data 6.