Introduction

Oral squamous cell carcinoma (OSCC) is one of the most common tumor entities worldwide, with approximately 600,000 new cases per year and a cumulative 5-year survival rate of only 55–60 % [1]. In recent years, the poor prognosis has hardly changed, despite the introduction of new therapeutic and diagnostic procedures [1, 2]. In Germany, the annual incidence is approximately 13,800 new cases per year and increasing [3]. In the field of adjuvant drug tumor therapy, there has also been almost no change except for the 2006 approval of the epidermal growth factor receptor (EGFR) antibody cetuximab, which is highly controversial because of its limited clinical efficacy [3, 4, 5]. However, significant innovations have been shown in drug tumor therapy of other tumor entities. In addition to classical cytostatic therapy, the so-called targeted therapy, including the above-mentioned monoclonal antibodies or tyrosine kinase inhibitors (TKI), has found its way into the therapy of other tumor entities, such as non-small cell lung cancer (NSCLC) or malignant melanoma. Another promising approach in drug tumor therapy is the attempt to overcome the resistance of tumor cells to cell death signaling. Apoptosis is induced by ligands of the tumor necrosis factor receptor (TNFR) superfamily, such as CD95 and TNF receptor [6]. This extrinsic pathway is used to detect apoptotic signals from the direct cellular environment. Transmembrane receptors trimerize by linking the death ligands (e.g., FasL, TRAIL, TNF-α) to their extracellular portion, which leads to apoptosis by activating the caspase cascade [7, 8]. Via nuclear stress, an intrinsic pathway is induced and results in a change of permeability of the mitochondrial outer membrane and thus in a release of second mitochondria-derived activator of caspase/direct IAP-binding protein with low PI (SMAC/DIABLO), cytochrome C, endonuclease G, and apoptosis-inducing factor (AIF), and high temperature required protein A2 (Htr A2/Omi) from the mitochondria [9].

Key proteins of apoptosis, so-called inhibitor of apoptosis proteins (IAPs), can counteract the activation and the activity of caspases [10]. They are often overexpressed in solid tumors of the head and neck and contribute to lymph node metastasis and a poor patient prognosis in HNSCC [11, 12, 13, 14]. cIAP-1 and cIAP-2 promote activation of NF-kappaB, which leads to the activation of inflammation and blockage of apoptosis. Pro-apoptotic molecules, including SMAC, secreted by the mitochondria act as potent inhibitors of this IAP family and induce apoptosis again. LCL-161 is a potent inhibitor of cIAP1 and XIAP that initiates their destruction and induces apoptosis via caspase activation [15]. Promising results from a combination of inhibition of XIAP, survivin, and EGFR in the therapy of NSCLC were shown by Lee and colleagues. Yang et al. reported a significantly greater reduction in cell count in in vitro studies in combination therapy with radiation therapy and a SMAC-mimetic compound (SM-164). Synergistic effects in combination therapy with SMAC-mimetic compounds and TRAIL in gliomas and neuroblastomas, melanoma, pancreatic carcinoma, and leukemia were shown recently.

This underlines the urgent need to explore further opportunities for drug tumor therapy to improve the still-poor prognosis of HNSCC. This study aims to investigate the efficacy of the SMAC-mimetic compound LCL161 in combination therapy with FasL in the treatment of HNSCC. To our knowledge, this is the first study to investigate the efficacy of LCL161 in combination therapy with FasL in the treatment of head and neck squamous cell carcinoma.

Material and methods

Cell lines

The cell culture was carried out in a sterile safety cabinet. All cell lines were cultured in a gassing incubator at 37 °C and 5 % CO2. Media were changed four times a week. The medium used for this purpose was Dulbecco’s modified Eagle’s medium (DMEM), low glucose (Invitrogen, Karlsruhe, Germany; 4.5 g/L D-glucose, 4 mM l-glutamine, 110 mg/L sodium pyruvate) with 10 % fetal calf serum (FCS) and 10,000 U/ml penicillin-streptomycin (Life Technologies, Darmstadt, Germany) added. The cells were detached by 0.25 % trypsin and 0.53 mM EDTA from the breeding bottle. Cell count was carried out by a volume-based aggregation measurement (Casy, Rack Systems, Schärfe Systems, Penzberg, Germany). This was followed by seeding the cells in 96-well plates. Each well was loaded with 100 μl medium with 10,000 cells. After re-incubation for 24 h and adhesion of the cells at the bottom of the wells, the cells were incubated with the different compounds. All experiments were performed at least in triplicate, and the average value was used to evaluate the results.

Treatment with Fas ligand and LCL161

FasL and LCL161 were purchased from Selleckchem (distributed by Absource Diagnostics GmbH, München, Germany) and stored according to the manufacturer’s protocol. The FasL concentrations (200, 100, 50, 25, 12.5, 6.025, 3.125, 1.5625 μM) and the LCL161 concentrations (200, 100, 50, 25, 12.5, 6.025, 3.125, 1.5625 μM) were derived from a log2 dilution. These concentrations are based on findings of Tian and colleagues. They described the effect of LCL161 in the treatment of hepatocellular carcinoma cell lines in vitro [16]. For our investigations in combination treatment, a constant concentration of LCL161 referring to the IC50 of each cell line was used (Table 2). After addition of the drugs, the cultures were incubated for another 72 h.

Crystal violet assay

Cytotoxicity was measured by the crystal violet test, a spectrophotometric method, which is based on the dyeability of cells. This test allows cells to be counted by measuring their absorption. After 72 h of incubation, the medium was removed with a micropipette and the remaining cells in the 96-well plate were stained for 12 min with crystal violet and then washed with distilled water several times. The plates were dried for 24 h and measured for absorption. For this purpose, 100 μl of methanol was added to each well and the cells were measured photometrically after a 10-min exposure at 595 nm wavelength in an ELISA reader (Rainbow Spectra, Tecan, Mannedorf, Switzerland).

FACS analysis

CD95 expression in the several tumor cell lines was detected with fluorescence-activated cell sorting (FACS). For this purpose, CD95/Fas antibodies were used to label the receptors. IgG2B isotype control PE antibody (R&D Systems, Minneapolis, MN, USA) was used as the negative control. The reagents were used according to the protocols described by the manufacturers.

Western blot

Cell lysates of the above-mentioned tumor cell lines were scraped into ice-cold PBS with a rubber policeman, collected by 10 min centrifugation, and lysed after sonification (10 pulses) by boiling (5 min, 96 °C) in 4× Laemmli sample buffer (8 % SDS, 0.1 M DTT, 40 % glycerol, 0.2 M Tris, pH 8.0) supplemented with phosphatase inhibitor cocktails I and II (Sigma-Aldrich, St. Louis, MO, USA). Proteins were separated by 8 % SDS-PAGE and transferred to nitrocellulose membranes as described by Laemmli 1970 [17]. Immunoblotting was performed with primary antibodies specific to XIAP, cIAP1, and cIAP2 (Cell Signaling Technology, Danvers, MA, USA).

Statistical analysis

Statistical analysis of the data was performed with Microsoft Excel 2007 (Microsoft, Redmond, WA, USA) and Prism 6.04 (GraphPad, La Jolla, CA, USA). Two statistical aspects were investigated: First, we investigated the different efficacies by incubating the single agents LCL161 and FasL. Second, we examined the use of a combination with these two substances in every cell line. After at least three repeats of the experiments, data were evaluated with a nonparametric Mann-Whitney test. The significance level was set at p ≤ 0.05. The statistical analysis is based on several repeated and representative experiments.

Results

FACS analysis

CD95/Fas receptor was detected via flow cytometry in every cell line (PCI-1, PCI-9, PCI-13, PCI-52, and PCI-68) (Fig. 1).

Fig. 1
figure 1

Results of the FACS analysis. CD95 (Fas ligand receptor) was detected in all cell lines

Full size image

Western blot

Under control conditions, cIAP1 was clearly expressed in every cell line (PCI-1, PCI-9, PCI-13, PCI-52, and PCI-68). After the addition of LCL161, degradation of cIAP1 was detected in all cell lines by Western blot analysis (Fig. 2).

Fig. 2
figure 2

Results of the Western blot analysis showing a distinct degradation of cIAP1 after treatment with LCL161

Full size image

Treatment efficacy in PCI-1 cells

PCI-1 cells showed a concentration-dependent responsiveness to the different concentrations of FasL (log2 dilution). The control group was set to 100 % (SD ±4.9 %). The viable fraction for the FasL concentration of 2 μM was 84.5 % (SD ±3.3 %). In response to FasL 3 and 6 μM, a viable fraction of 72.7 % (SD ±2.7 %) and 59.6 % (SD ±1.5 %) was shown. After treating the cells with a FasL concentration of 13 μM, cell viability fell to 44.4 % (SD ±2.2 %). The viable fractions in the presence of the FasL concentrations 25 and 50 μM were 27.5 % (SD ±3.2 %) and 16.2 % (SD ±1.1 %), respectively. At 100 and 200 μM, cell viability fell to 16.5 % (SD ±1.5 %) and 20 % (SD ±2.9 %), respectively.

The cell lines showed varying reactions to the different concentrations of FasL (log2 dilution) in combination with a fixed concentration of LCL161 (IC50, 12 μM). The viable fraction in the presence of FasL 2 μM in addition to the fixed concentration of LCL161 was 37.6 % (SD ±6.6 %). FasL 3 and 6 μM combined with LCL161 caused a cell viability of 24.8 % (SD ±5.2 %) and 13.6 % (SD ±3.8 %), respectively. The combination of FasL 13 μM and LCL161 reduced cell viability to 11 % (SD ±2 %). A reduction to 9.5 % (SD ±1.9 %) and 8.9 % (SD ±1.3 %) was detected for FasL 25 and 50 μM with LCL161. The combination of FasL 100 or 200 μM with LCL161 reduced cell viability to 9.7 % (SD ±0.6 %) or 10.2 % (SD ±0.5 %), respectively. The comparison of the individual concentrations in single and in combination showed highly significant differences (p < 0.0022) (Table 1; Fig. 3a).

Table 1 Name, origin, and TNM status of the five cell lines used in the study
Full size table
Fig. 3
figure 3

Viable fraction of the cell culture as measured by the crystal violet assay. As illustrated by the black columns, FasL monotherapy resulted in a concentration-dependent decrease of the viable fraction in cell lines PCI-1, PCI-13, and PC-68 (a, c, e). In contrast, FasL monotherapy did not have a relevant cytotoxicity in PCI-9 and PCI-52 cells (b and d). In four out of five cell lines (PCI-1, PCI-9, PCI-13, and PCI-68), combination treatment with FasL and LCL161 at the IC50 concentration (Table 2) led to significant enhancement of treatment effects (a, b, c, e). However, PCI-52 remained resistant to FasL even if coincubated with LCL161 (d). In this case, the decreased viable fraction in the presence of combination treatment seemed to be triggered by LCL161 alone. The error bars indicate the standard deviation

Full size image

Treatment efficacy in PCI-9 cells

PCI-9 cells were nearly resistant to FasL in log2 dilutions. The control was set to 100 % (SD ±4.8 %). FasL 2 μM reduced cell viability to 94.6 % (SD ±4.8 %). The concentrations of 3 and 6 μM showed a little effect on the cell line [96.2 % (SD ±6.9 %) and 101.8 % (SD ±7.8 %) viability, respectively]. A viable fraction of 100.4 % (SD ±6.5 %) was shown for the FasL concentration of 13 μM. Cell viability decreased to 89.5 % (SD ±4.9 %) and 91.4 % (SD ±4.4 %) with FasL 25 and 50 μM, respectively. The viable fractions in the presence of 100 and 200 μM were 98.2 % (SD ±4.7 %) and 101.5 % (SD ±3.7 %), respectively. Except for the comparisons of 13 vs. 25 μM and 50 vs. 100 μM, there were no significant differences in viability (Table 2).

Table 2 IC50 of LCL161 in the different cell lines
Full size table

The PCI-9 cells responded differently to combination treatment. FasL was added in log2 dilutions and LCL161 in a fixed concentration (IC50 of 26 μM). A viable fraction of 39.12 % (SD ±13.8 %) and 37.9 % (SD ±11.3 %), respectively, was shown for the combination consisting of FasL 2 μM and FasL 3 μM with LCL161. FasL 6 and 13 μM in combination with LCL161 reduced cell viability to 31.6 % (SD ±11.4 %) and 27.4 % (SD ±10 %), respectively. FasL 25 μM and LCL161 caused a viable fraction of 25.1 % (SD ±8.2 %). FasL 50 μM and LCL161 caused a viable fraction of 28.1 % (SD ±7.1 %). FasL 100 and 200 μM in combination with LCL161 reduced cell viability to 29.4 % (SD ±7.5 %) and 27.5 % (SD ±6.2 %), respectively. Comparing the individual concentrations of mono- and combination treatment, highly significant differences were seen (p < 0.0022) (Tables 1 and 3; Fig. 3b).

Table 3 Results of the statistical analysis
Full size table

Treatment efficacy in PCI-13 cells

PCI-13 cells showed varying responsiveness to FasL. The control was set to 100 % (SD ±6.8 %). The FasL concentrations of 2, 3, and 6 μM showed no significant effects on cell viability (range, 99.6 to 97.6 %). A viable fraction of 74.7 % (SD ±4.1 %) was detected for the FasL concentration of 13 μM. Cell viabilities of 38.6 % (SD ±3.2 %) and 20.4 % (SD ±4.5 %) were shown for the FasL concentrations of 25 and 50 μM, respectively. FasL 100 and 200 μM caused a viable fraction of 11.9 % (SD ±1.7 %) and 21.8 % (SD ±0.9 %), respectively.

Synergistic effects were seen when incubating both agents in combination. The viable fraction of 24.8 % (SD ±1.4 %) and 18.6 % (SD ±2.2 %) was detected for FasL 3 and 6 μM with a fixed concentration of LCL161 (IC50 of 11 μM). Cell viability decreased to 18.6 % (SD ±2.2 %) and 12.9 % (SD ±0.7 %) in the presence of FasL 6 and 13 μM in combination with LCL161. For the combinations of FasL 25 and 50 μM with LCL161, a viable fraction of 9.9 % (SD ±0.9 %) and 7.9 % (SD ±1.9), respectively, was detected. Viabilities of 9.8 % (SD ±2.9 %) and 8.7 % (SD ±0.8 %) were shown for the combinations of FasL 100 and 200 μM with LCL161. Comparing the results of the incubation with the single agent and the combination of both agents, we saw a highly significant (p < 0.0022) difference in cell reduction in the combination treatment, except at the FasL concentration of 100 μM (Table 1; Fig. 3c).

Treatment efficacy in PCI-52 cells

PCI-52 cells showed little response to the varying concentrations of FasL (2–200 μM). Cell viability ranged from 96.9 to 113.3 %. In combination with FasL and the IC50 concentration of LCL161 (13 μM), PCI-52 cells also showed little effect (Table 1; Fig. 3d).

Treatment efficacy in PCI-68

The cell line PCI-68 showed a heterogeneous vulnerability towards FasL. We set the control group to 100 % (SD ±1.4 %). A viable fraction of 99.7 % (SD ±1.8), 100.1 % (SD ±1.6 %), and 100.2 % (SD ±1.5 %) was detected for FasL concentrations 2, 3, and 6 μM, which were all similar to control (p > 0.5). FasL 13 and 25 μM reduced the viability to 96.9 % (SD ±1.7 %) and 76.7 % (SD ±3.1 %), respectively. A viable fraction of 53.7 % (SD ±4.2 %), 37.9 % (SD ±1.5 %), and 28.5 % (SD ±1.2) was detected for the FasL concentrations 50, 100, and 200 μM, respectively. The last five concentration increases showed highly significant (p < 0.01) differences between each other.

Varied reactions were detected in the combination with FasL (log2 dilution) and the referring LCL161 concentration (IC50, 17 μM). A viable fraction of 72.6 to 83 % was shown for the FasL concentrations 2, 3, and 6 μM combined with LCL161 17 μM. FasL 13 and 25 μM plus LCL161 caused a viable fraction of 58.9 % (SD ±3.2) and 40.1 % (SD ±2.9 %), respectively. A viable fraction of 30.5 to 21.5 % was detected for the FasL concentrations 50, 100, and 200 μM with a fixed concentration of LCL161. The comparison of the individual FasL concentrations in mono- and in combination treatment showed highly significant differences (p < 0.0022) (Table 1; Fig. 3e).

Discussion

Oral squamous cell carcinoma is one of the most common malignancies worldwide. HNSCC patients are typically treated with a multimodality approach of systemic chemotherapy, radiation, and surgery. Cetuximab, a monoclonal antibody directed against EGFR, is the only FDA-approved molecular agent for this disease. However, cetuximab has demonstrated limited efficacy, where both intrinsic and acquired resistance to this therapy is a common outcome. Thus, identifying new molecular targets in this disease is critical. Understanding how HNSCCs bypass apoptotic stimuli may identify new avenues for therapeutic intervention in this disease. Apoptosis can be induced by both cell intrinsic and extrinsic stimuli. The cell intrinsic pathway can be activated by DNA damage or ER stress, which will change the permeability of the mitochondrial outer membrane leading to the release of SMAC/DIABLO and subsequent caspase activation. SMAC is a negative regulator of the IAP family [18, 19], which are often overexpressed in solid tumors including HNSCC. Thus, SMAC mimetics are now being considered as therapeutic agents in oncology [20–22].

Other studies have shown promising effects of SMAC mimetics in preclinical models of HNSCC. These mimetics bypass NF-κB-mediated cell survival pathways, enhance radiosensitivity, and improve the efficacy of other therapeutics used to treat HNSCC. Based on these findings, we aimed to determine if the SMAC mimetic, LCL161, could enhance the effect of FasL in HNSCC preclinical models.

Equipping the cells with FasL, we found two of five cancer cell lines to be resistant to FasL. We observed heterogeneous responses to LCL161 (used at different IC50 values) and the combination with FasL and LCL161. Three of the five cell lines were sensitive to FasL. The combination treatment with LCL161 had additive effects, as in PCI-68 cells. Synergistic effects were observed with the combination therapy in the cell lines PCI-1 and PCI-13.

Interestingly, apoptosis could be induced in the FasL-resistant cell line (PCI-9) when co-treated with LCL161. These data suggest that the IAP family members XIAP, cIAP1, and cIAP2 may be responsible for this cell line’s resistance to FasL. This finding is also supported by Western blot analysis demonstrating cIAP1 degradation after LCL161 treatment in all cell lines. Furthermore, we found a pronounced induction of apoptosis with the smallest concentration range, which underlines the above-mentioned hypothesis. Our finding that LCL161 induced apoptosis when combined with FasL is consistent with a study that investigated the efficacy of a combination therapy with TNF-α, TRAIL, and SM-164 [23]. This finding is also supported by other studies demonstrating a significant sensitization of HNSCC cells to gemcitabine in combination with a SMAC mimetic.

Only one cell line (PCI-52) did not respond to the combination. We detected a significantly lower cell concentration in the combination therapy compared to the monotherapy, but we expect there is no clinical relevance in these findings. PCI-52 cells may be resistant to combinatorial therapy because they overexpress several IAP family members. PCI-52 cells in general seem to be equipped with many anti-apoptotic mechanisms as demonstrated by comparing the sensitivity of PCI-52 cells with other cell lines treated with cisplatin, 5-fluorouracil, docetaxel, paclitaxel, cetuximab, panitumumab, erlotinib, and gefitinib [24, 25]. PCI-52 cells are largely resistant to these therapies when tested in vitro.

This is the first study to demonstrate the enhanced apoptotic effect of combinatorial FasL and LCL161 treatment in HNSCC preclinical models. In the majority of the cell lines, apoptotic effects induced by FasL were enhanced by the addition of LCL161. Notably, one out of two FasL-resistant cell lines (PCI-9) was sensitized to FasL treatment when co-treated with LCL161. Collectively, these findings provide rationale for the use of SMAC mimetics in combination with prop-apoptotic therapies for the treatment of HNSCC and demonstrate the importance of co-targeting several anti-apoptotic signaling nodes to achieve maximal benefit.

Although these findings are encouraging, the results cannot be extrapolated into a clinical situation. All findings were made in an in vitro setting and further investigation is necessary before trying to translate them into clinical implications.