Head and neck squamous cell carcinomas (HNSCC), including cancers of the oral cavity, pharynx and larynx, account for approximately 5% of all cancers worldwide [1]. Primary surgical eradication of the tumor and removal of the regional lymph nodes is the first line therapy in many cases. Subsequently, depending on the pathological staging, a number of patients are eligible for adjuvant treatment with either radiotherapy (RT) or radiochemotherapy (RCT). Especially patients with locally advanced HNSCC, involved resection margins and/or extranodal extension of cancer growth in cervical lymph node metastases should receive concurrent chemoradiation. Studies have shown a local recurrence of 27%-61% as well as regional metastasis as high as 21% and a 5-year survival of 27%—34%, after surgery and adjuvant radiation [2]. However, the use of cisplatin in the standard dose and regime of 20-70 mg/m2 every 3 weeks is associated with severe toxicities such as renal- and ototoxicity, and myelosuppression [3, 4]. Numerous HNSCC show intrinsic resistance to platinum drugs [5]. Moreover, many cancers tend to develop multidrug resistance in the course of treatment [6]. Severe side effects of cytostatic treatment frequently lead to incompliance, discontinuation, treatment failure and incompletion of the planned therapy, which increases the risk of cancer recurrence [6]. Novel results hint that low intracellular ROS levels may be the decisive step in cisplatin resistance of cells. In order to decrease side effects, increase therapy efficacy and improve the overall survival rate, recent studies have focused on comparing high-dose cisplatin with cisplatin based combination therapies [7,8,9].
Cold atmospheric plasma (CAP) is a promising alternative and additional treatment to the current cancer therapies [10, 11]. CAPs are partially ionized gases, which can be generated at atmospheric pressure and operate under room temperature. In this study direct plasma application was used. Both physical and chemical factors in direct CAP treatment have been shown to have an impact on malignant cell viability reduction. Chemical effects have been shown to have the main influence on viability reduction in in-vitro studies. Common cellular responses include the rise of intracellular ROS, DNA damage, as well as mitochondrial and cytoplasmic membrane damage. Several studies have revealed that H202 and NO2 are the main reagents associated with CAP exposure of cells. The physical effects include thermal ultraviolet irradiation, and electromagnetic effect and have been shown to be minor in CAP in-vitro application in previous studies [12, 13]. Studies have not only proven the anti-microbial efficacy of CAP on human skin and mucosa, but have also proven to be a promising application in the treatment for HNSCC [11, 14,15,16]. CAP has shown to induce cellular responses such as cell apoptosis, inhibition of growth, selective cancer cell death, DNA damage and/or cell cycle arrest, in this respect being more effective and less toxic than some other common therapies such as radiation and chemotherapeutics in cell line experiments [12, 14]. The toxic effects of CAP on healthy tissues are minor compared to its strong impact on cancer cells in vitro [17, 18].
Unlike many other tumors, HNSCC can be accessed directly via the oral orifice. Thus, a direct CAP application in-vivo on the tumor would theoretically be possible. Due to the previously mentioned toxicity of cisplatin, an increase of the systemically applied dose is not possible in extensive or recurrent HNSCC. Therefore, the combination of CAP and cisplatin application could be of clinical relevance. In this study the common cellular responses were detected and analysed by comet assay (DNA damage), MTT assay (mitochondrial damage) and trypan blue staining (cytoplasmic membrane damage). Physical factors of CAP exposure have been shown to be minor in CAP in-vitro application in previous studies. The culture temperature of the cell medium in our study was set at 37 °C. Studies have shown that even close CAP exposure will not increase the temperature significantly. Furthermore, the ultraviolet light exposure of cancer cells fails to show significant antiproliferative effects on cancer cells in-vitro and therefore physical factors were not specifically addressed in this study [19, 20].
The purpose of the present study was to investigate the effect of CAP treatment (exposure times 30 s, 60 s, 90 s, 120 s and 180 s) on common HNSCC cell lines (FaDu and OSC-19) in comparison to a chemo-resistant HNSCC cell line (Cal 27). A further aim was to examine the therapeutic efficacy of low dose cisplatin in combination with CAP for cisplatin sensitive and resistant HNSCC cell lines in-vitro.
Materials and methods
Cell culture
Three different HNSCC cell lines were used for this study. The same two cell lines (OSC 19 (JCBR Cell BANK) and FaDu (ATCC, Manassas, VA, USA)) as in the study of Welz et al. were used [17, 21] and Cal 27 cell line (ATCC, Manassas, VA, USA), an oral adenosquamous carcinoma cell line which is known to be resistant to cis-platinum.
OSC 19 cells were grown in DMEM/Ham´s F-12 (Biochrom AG, Berlin, Germany) and FaDu cells in DMEM (Biochrom). Both cell lines were also supplemented with 10% fetal bovine serum (Gibco, Carlsbad, CA, USA) and 10 U/ml Penicillin–Streptomycin (Biochrom). FaDu cells were additionally supplemented with 1% non-essential amino acids (Invitrogen, Karlsruhe, Germany) and each 1% of L-Glutamine and Sodium Pyruvate (Biochrom). Cal 27 cells were cultured in DMEM also supplemented with 10% fetal bovine serum (Gibco, Carlsbad, CA, USA) and 10 U/ml Penicillin–Streptomycin (Biochrom). All cells were preserved under humidified conditions with 5% CO2 at a temperature of 37 °C. All cells were cultured in 25 ml culture flasks (Nunc EasYFlasks) and medium was changed every 2–3 days. For this study, 5 × 105 cells of Cal 27, FaDu or OSC 19 were seeded onto a 6 well plate, and were left to attach for 16 h. For the experiments the cells seeded were splitted at least 3 times.
Plasma treatment
A CAP device using the surface microdischarge (SMD) technology for plasma production was used for this study [22]. An SMD device is a modified dielectric barrier discharge (DBD) device. The MiniFlatPlaSter® device details, which was used, are described in Maisch et al. [23], Welz et al.[18] and Welz et al. [17] and visualized in Figs. 1, 2 and 3.
A glass epoxy board, which is sandwiched by a stainless-steel mesh grid and copper foil layer creates the SMD electrode. The plasma is produced homogenously on the mesh grid side in the air by applying a high pulse-like voltage of 7 kV with a repetition frequency of 6.75 kHz. Via the mechanism of diffusion, the generated reactive species are transferred to the cells. The MiniFlatPlaSter® produces the plasma indirectly.
One well of a 6-well plate has a rim diameter of 28 mm, fitting the CAP electrode exactly. The medium was then completely removed and the cells were exposed to CAP treatment. The CAP device was placed exactly onto the well rim, causing a closed volume condition (distance between the electrode and cells was 17.5 ± 0.5 mm). The respective controls were treated equally as the plasma treatments except for the CAP exposure. The treatment times used were 30 s, 60 s, 90 s, 120 s and 180 s. Immediately after the appropriate treatment medium was added.
Cisplatin treatment
Ten µM cisplatin have been shown to induce sufficient DNA cross-links without inducing apoptosis in cell cultures [21]. The cisplatin concentration of 2.5 µM, which was used in this study, was chosen after analysing the dose–response curve (Fig. 4) of all three different HNSCC cell lines. A low dose of cisplatin was desired with barely any impact on the different HNSCC cells in order to differentiate if a combination therapy could have the same or even greater impact on the cell viability with less cytotoxicity through chemotherapy. Dimethyl sulfoxide (0.05%) (DMSO; Merck, Darmstadt, Germany) was used as a solvent for cisplatin solution preparation. The solution was prepared fresh and protected from light right before the experiments.
Prior to cisplatin application, all cells were seeded and exposed to CAP treatment as stated above in the method and materials section. After the described CAP exposure each cell type was divided into another two treatment groups. For both groups the cells were incubated with 1 ml trypsin/EDTA solution for 8–10 min for trypsination and seeded at 8000 cells/well (100 μl) in a 96 well plate. The first group, cisplatin concurrent (Cis + cc) underwent a 2.5 µM cisplatin application directly after the different CAP treatments. In the second group, cisplatin consecutive (Cis + cs), the cells were incubated (37 °C with 5% CO2), in the 96 well plates in their according medium after CAP treatment and underwent a 2.5 µM cisplatin application after 24 h of incubation (37 °C with 5% CO2). For both groups 2.5 µM of cisplatin was added to each 96 well plate at the stated time point and the well plates were incubated under the same conditions as stated before for another 24 h. Following this, cisplatin was removed and each well plate was washed with PBS. PBS was then discarded and replaced with 10 μl labeling medium containing 0,5 mg/ml MTT and an MTT assay as described in detail below was performed.
Cell viability/trypan blue staining (exclusion test)
Trypan blue is used to distinguish between viable and dead cells. The method is based on the principle that the dye is not absorbed by living cells with an intact cell membrane. Dead cells take up the dye due to their damaged membrane and are stained blue [24]. To analyse the cell viability changes after the different CAP treatment times for the different cell types, the trypan blue exclusion test was performed as described by Welz et al. [18]. This was performed for each cell type after CAP treatment (30 s, 60 s, 90 s, 120 s and 180 s). The cells were separated and PBS was discarded after centrifugation (800 rpm for 5 min at 4 ◦ C). The pellets of the treated cells were resuspended in 1 ml of 1 × PBS. 50 μl of cell suspension was then mixed with an equal volume of trypan blue 0.4% (Merck) and transferred to a hemocytometer slide. At least 200 cells were metered for each data point in sixteen microscopic fields. Following that the cells were counted using a light microscope. The percentage of viable cells = ((non-stained cells) / (stained + non-stained cells)) × 100 [17, 18].
Cell viability/MTT asssay
The MTT Assay is used to assess cytotoxicity, proliferation and cell viability by measuring the cellular metabolic activity [25]. In this study, the Cell Proliferation Kit I Roche Diagnostics (Roche Diagnostics GmbH, Mannheim, Germany) was used according to the instruction manual as described by Welz et al. [17]. To monitor changes in cell viability after different treatment modalities (CAP, CAP + cc cisplatin application and CAP + cs cisplatin application) and CAP exposure times, the treated cells and the respective controls were trypsinized and seeded at 8000 cells/well (100 μl) in a 96 well plate. The metabolic cell activity was measured 24 h after CAP treatment or CAP + cc cisplatin application or CAP + cs cisplatin application. The culture medium was replaced with 10 μl labeling medium containing 0,5 mg/ml MTT. After 4 h of incubation in a humidified atmosphere (37 °C with 5% CO2), 100 μl MTT-staining solution was added to each well followed by an incubation overnight. A VERSAmax™ ELISA- Reader (Molecular Devices GmbH, Biberach, Germany) at a wavelength of 550 nm was used to quantify the purple formazan dye. The reference wavelength corresponds to 690 nm [17, 18]. Every experiment contained triplicate measurements of cell viability reduction of each plasma treatment time and each was repeated five times (measurements for each treatment time n = 15). The mean cell viability reduction curves were standardized to the percentage of living cells, whereas the control cells were set at 100%.
DNA damage/alkaline microgel electrophoresis (comet assay)
For the detection of DNA damages, the alkaline microgel electrophoresis (Comet assay) was performed after the different CAP treatments. Depending on the extent of damage/fragmentation, the DNA varies in its migration behaviour in an electrical field. Undamaged DNA shows no migration in comparison to fragmented DNA. The higher the fragmentation (damage) the further and faster the migration [26, 27].
The alkaline microgel electrophoresis was carried out as published in the methods and materials section of Welz et. al. [17].
After the CAP treatment the cells were incubated with 1 ml trypsin/EDTA solution for 8–10 min for trypsination. Following, neutralization, centrifugation (10 min, 900 U/min), cell counting and cell viability screen with a trypan blue exclusion test were carried out. A two layer agarose was used to ensure stability. The cells were resuspended in 75 µl of 0.7% low-melting agarose (Biozym, Hameln, Germany), applied to slides (Langenbrinck, Emmendingen, Germany) which were covered with normal melting agarose (Biozym) to ensure the stability. The slides were then immersed with alkali solution for 1 h (10% DMSO, 1% Triton-X, 2.5 M NaCl, 10 mM Trizma-Base, 100 mM Na2 EDTA and 1% N -lauroylsarcosine sodium salt). After the lysis process, the slides were placed in the gel electrophoresis chamber (Renner, Dannstadt, Germany). Before applying an electric field, they were left with alkaline buffer solution containing 300 mMNaOH and 1 mM Na2EDTA at pH 13.2 for 20 min for the DNA double helix to denature. The electrophoresis was started at 0.8 V cm − 1 and 300 mA and continued for 20 min. The slides were then neutralised with Trisma base, 400 mM, pH 7.5 (Merck, Germany). After this, they were stained with 75 μl ethidium bromide (Sigma; [51 μM]) and analysed with a DMLB microscope (Leica, Bensheim, Germany). By random pattern, 80 cell nuclei per slide (2 slides per CAP treatment time) were selected and digitized with the attached monochrome CCD camera (Cohu Inc., San Diego, CA, USA). Using the image analysis software Komet + + (Kinetic Imaging, Liverpool, UK) DNA-migration was measured. Cells containing damaged DNA have the appearance of a comet under the microscope, where the undamaged DNA is the „head “ and the damaged DNA is the „tail “ of the comet (Fig. 1 (a-b)). Intact DNA is shown as an intact nucleus, „head “, with no tail. For orientation of genotoxicity the Olive Tail Moment (OTM) was used and automatically calculated by the computer software (OTM = (tail mean-head mean) x % of DNA in the tail) [28]. Cells which showed an OTM < 2 were considered as undamaged [29]. The migration was measured by the software and calculated by the % of DNA tail. This is the relative fluorescence intensity in the “head” and “comet tail” [30] (Fig. 5).
Apoptosis/Annexin V Fit-C
Annexin V Fit-C is a rapid and sensitive method to determine apoptotic from necrotic cells. 24 h after CAP treatment the induction of apoptosis was investigated with fluorescence microscopy using an Annexin V Fit-C detection kit (PromoKine, Heidelberg). CAP treatment and trypsinisation were carried out as described above. HNSCC cell lines were stained with Annexin V Fit-C and propidium iodide (PI) according to the manufacturer’s instructions. In this case, 1 × 105 cells were resuspended in 500 μl binding buffer solution, incubated for 5 min under red light with 5 μl Annexin V-FITC and 5 μl propidium iodid (PI). Phosphatidylserin (PS) residues are found on the inner surface of the membrane in normal cells, hence being inaccessible to Annexin V. An early step of apoptosis is the translocation of phosphatidylserin (PS) from the internal to the external face of the plasma membrane as the cell membrane becomes permeable. Annexin-V is a Ca 2+ phospholipid binding protein with a high affinity to PS and can therefore be used as a sensitive marker for the apoptotic cells. Combining Annexin-V labelling with PI staining allows discriminating between (early) apoptotic and late apoptotic/necrotic cells, as PI will be able to stain the DNA in the nucleus. One hundred cells per CAP treatment were counted with the help of a hemocytometer and early apoptotic cells were identified with fluorescence microscopy, and indexes calculated out of 100 cells (Fig. 6).
Statistical analysis
Unless stated otherwise, the data is presented as the arithmetic mean ± 95% CI. Graph Pad Prism 7.0. Software was used for statistical analysis. The Two-way ANOVA test with a Bonferroni correction test as a post-test, to counteract the problem of multiple comparisons, was used to calculate the statistical significance of the results. Differences were considered as significant at the calculated stated adjusted p-values, prior the statistical analysis.